Pigeons and Pickle Jars

By Jonah Goldberg

Wednesday, May 27, 2026

 

A long time ago, I read The Africans by veteran foreign correspondent David Lamb. He recounted a tendency for African drivers to draw suboptimal conclusions from near misses. He wrote:

 

It is difficult to find a satisfactory explanation for the Africans’ propensity to pass on blind curves and drive at out-of-control speeds. The best one, I suppose, is that an African does not conceptualize a potential problem the way a Westerner does. The Westerner says, If I do this, that might happen. The uneducated African does A without reasoning that it could lead to B. If an oncoming car has to swerve off the road to avoid his vehicle, and there is no collision and no injuries, the African does not say, Next time I’d better not do that. He will do exactly the same thing because he has, after all, accomplished his objective of getting from one point to another without major mishap. He does not deal with the unexpected on a sophisticated level because to do so is, again, a quality of education and training, and the automobile is a new device to most Africans.

 

I’ve heard similar observations about drivers in South America and parts of Asia. I’ve had conversations about this phenomenon for years, but I’ve grown uncomfortable with it. Lamb’s sweeping generalization about a vast, diverse, continent—not to mention throwing in an even bigger chunk of the global population—is part of my discomfort. But the bigger portion is that I don’t think the tendency he identifies is limited to the uneducated, never mind the nonwhite uneducated.

 

One of the greatest, and most human, of mistakes is to think if an idea worked, it was a good idea.  Intellectuals, including intellectuals of the whitest shades of pale, make this mistake all of the time. 

 

Soviet agronomist Trofim Lysenko succeeded at germinating seeds with extra moisture and cold to produce earlier flowering and assumed this strategy would work in every conceivable context, including engineering superior humans. His crackpottery ended up getting a lot of people killed.

 

Henry Heimlich invented the Heimlich maneuver, which really worked and saved countless lives. But he didn’t think it should be reserved for choking victims. He and his supporters advocated for it to be used on drowning victims, asthma sufferers, and even people having heart attacks. This was a near perfect example of thinking that a good idea in one context must have applicability in very different ones.

 

The history of war is often the history of thinking past successes are predictive of future victories. Napoleon thought Russia was just the next trophy. Hitler had similar delusions. The Japanese had used surprise attacks or “decisive” early battles to great effect against Russia and China, so why wouldn’t it work with America at Pearl Harbor? This folly was dubbed “victory disease.”

 

I am barely scratching the surface of the myriad ways our brains trick us like this. Indeed, our brains’ bags of tricks are so full that psychologists, economists, statisticians, and of course gamblers have scads of terms to describe these kinds of errors, starting of course with post hoc, ergo propter hoc. Here are a few examples of such cognitive legerdemain: “Superstitious learning,” “winner’s bias,” “selection bias,” “resulting,” “omitted variable bias,” “base rate neglect,” “hindsight bias,” and the “hot hand fallacy.”

 

One of my favorites is the “Texas sharpshooter fallacy,” which can be neatly described as taking a bunch of bullet holes in the side of a barn and drawing bullseyes around them. This is a favorite technique of conspiracy theorists who impose conclusions retroactively on events—or Jews—to “prove” a theory.

 

So, it’s fine to dunk on bad drivers in Africa for their superstitious learning, but superstitious learning is also a term applied to CEOs who shepherd a merger at the beginning of an economic boom and conclude that their surging profits are the result of the merger, so they double down on the strategy. The voters who thought they’d be getting the pre-COVID economy of Donald Trump’s first term made the same error, thinking the president wills a good economy into existence. K Street is full of self-styled rainmakers who got rich by dancing when the first droplet hit the pavement so they could take credit for the downpour.

 

By the way, the term “superstitious learning” is rooted in the work of B.F. Skinner, who used the word “superstition” to describe how pigeons that bobbed their head before a pellet appeared assumed that the pellet resulted from their head bob. 

 

I’ve long argued that such biases are key to understanding Trump. He’s the millionth monkey banging on a type writer, or perhaps the one throwing darts at the stock pages. He ignores valuable and important rules and when things work out for him, he thinks the rules are B.S. His narcissism transmogrifies failure as the downpayment on success, and you can’t wholly blame him for it. But one proof that he’s an exception that proves the rule is that superstitious learners who imitate him tend to fail (unless he pardons them). This gives Trump’s luck too much credit, of course. He couldn’t run casinos for a profit. Casinos—the institutions that have most successfully monetized nearly all of the fallacies I’ve listed above.

 

But the guy was elected president twice (not three times) breaking all manner of rules. And that’s still pretty lucky.

 

Indeed, his war on Iran is a perfect example of how he is not immune to the rules, but thinks he is. He thought he had a hot hand after Venezuela and could do it again in Iran. But he couldn’t and now he’s trying to figure out how to say, “Never mind.” The Heimlich maneuver on Maduro yielded contrary results with the mullahs.

 

When being right is necessary but insufficient.

 

This was going to be the point of this “newsletter,” but I want to talk about a different kind of bias.

 

Kevin Williamson, as is his wont, said something interesting on the most recent episode of The Dispatch Podcast that got me thinking.

 

We were discussing the latest non-developments with the Iran war. Steve Hayes asked about the split on the right about to interpret how it’s been going. Part of Kevin’s explanation for why some Iran hawks were too quick to declare a kind of premature glide path to victory—if not outright victory—has to do with human nature.

 

“When you’re really, really dedicated to a particular policy, and you’re dedicated to it for a long time,” he said, you get “so used to talking about it in terms of [how] the benefits are really, really, big” and the “potential deficiencies and setbacks are and tradeoffs are really, really low because you’re used to being argumentative about it and that it’s hard to be analytic about it.”

 

Kevin admitted that he’d been thinking along these lines in the context of drug liberalization which—give some points to the anarchist-adjacent Kevin for honesty—hasn’t been going as great as some libertarians predicted.

 

This hit home for me in part because of all the policy disagreements I’ve had with friends I respect on the right—both conservatives and libertarians—drug legalization has probably been the most heated. I don’t want to debate the actual disagreement here, I want to discuss the tendency among champions of legalization and countless other policies to think that because they have won—by their lights—the theoretical debate that their policy preference is not merely right but the downsides are all negligible or even fictitious.

 

I remember arguing with a very respected libertarian intellectual about drug legalization. I warned that legalizing heroin would produce more heroin addicts. He responded, in part, by quoting Thomas Jefferson: “No nation is drunken where wine is cheap; and none sober, where the dearness of wine substitutes ardent spirits as the common beverage.” His point was that in a world where all drugs are legal, people will tend toward ones that they can live productively using. It was a version of an argument I made in college about how it was dumb to ban kegs in dorms, because that will encourage students to sneak in hard liquor and drink behind closed doors. Beer drinking in public is safer than Jim Beam shots in secret.

 

But, again, that’s not the relevant part. It was my libertarian friend’s profound, preemptive, ennui with my objections, which he’d heard a million times. He thought his answer was so obviously right and dispositive, it was almost an imposition for him to swat it away—to his satisfaction, not mine.

 

The truth remains that any effort to fully legalize heroin will create heroin addicts. We can debate whether it will create more or fewer addicts, but it won’t be zero. And I think a lot of libertarians, who extol human agency and freedom more than any other ideological group, are often smugly dismissive of the provable fact that drug addiction is evil precisely because it steals human agency and freedom for some fraction of humans. It’s a vampire problem they don’t want to fully acknowledge is a problem.

 

But this is only one potent example of this tendency. Socialists, of course, are a prime example of a group that lives in perpetual error but liberated from all doubt. And the history of undemocratic systems is in some respects the history of dumb ideas blowing up in the rulers’ faces—from Stalin’s collectivization of agriculture, to Mao’s Great Leap Forward, etc. But such errors in totalitarian systems are a different species of error.

 

In democracies the problem is different.

 

No, I have in mind serious people making serious arguments for good ideas. This, I think, explains why some Iran hawks misread Operation Epic Fury. I agree with them that regime change is the only real solution to the Iran problem. That doesn’t mean there wouldn’t be unintended consequences and Day 2 challenges. But the problem with Iran isn’t its nuclear program or ballistic missiles or its asymmetric leverage over the Strait of Hormuz. It’s the fanatics and thugs in power. When was the last time you worried about France’s nuclear weapons or Britain’s ballistic missiles?

 

I agree with those who’ve made arguments about the threat from the Iranian regime for decades. But winning debates about an idea in theory is, at best, half the job. 

 

When you think you’ve put the arguments to bed but haven’t gotten the policy you want, the energy of the cause moves toward someone with the will to implement your idea. We know what to do, we just need someone with the courage to do it. And because you’ve convinced yourself you’ve thought everything through, the executive’s ability to execute gets trimmed down to just a decision rather than a complicated project. After all, we’ve anticipated and defeated all the contrary arguments, so all we need is someone to make the decision and do it. It will be so easy.

 

Progressives are hardly immune to this problem, because it’s a problem endemic to intellectuals, to democracies, to politics, and to humans generally. How many progressives thought the “hard part” was getting California’s high-speed rail approved? It turns out that the hard part was building high-speed rail in a geographically bumpy, union-controlled, regulatory hellscape. It might have been hard to get the Affordable Care Act passed, but that difficulty pales in comparison to making healthcare, you know, affordable. But if you debated proponents of these programs before they were launched, the thing you probably remember most are the smug grins and eyerolls of policy geniuses who knew they were right.

 

I think we saw a similar dynamic with regard to Iran, because we’ve seen this throughout the Trump presidency, as Kevin noted. The immigration hawks, the government “efficiency” zealots, even the nationalist “intellectuals” felt that they handled the hard part by winning the argument to their own satisfaction. Indeed, often in the political realm, proof that they won the argument is that Trump got elected paying lip service to it. They loosened the jar of pickles. All that was left was the slightest effort to make it happen. So, when Trump picked up their respective pickle jars, they assumed they’d hear that thhhhffft of the vacuum seal breaking, a prelude to the sound of popping champagne corks and high fives.

 

The problem with the pickle jar analogy can be illustrated by changing it to almost any complicated problem where you still know you’re right. I can be absolutely sure that someone needs a heart transplant. Finding someone willing to do the operation isn’t nearly as important as finding someone capable of doing it.

 

It turned out that regime change, reinventing government with tech bro efficiency, removing illegal immigrants, or ushering in a nationalist restoration of the American Volksgemeinschaft requires more skill than opening a pickle jar.