There’s a time and place for evidence wars. I don’t think psychotherapy research, at its current stage of development, is one of them. The replication crisis has provided ample reason to be cautious about findings across the social sciences, and psychotherapy research has always been especially tricky.

The solution isn’t to abandon science, of course. It’s to do science better. Still, that doesn’t help a person right now. If, as Lee Jussim argues, ~75% of Psychology Claims Are False, how is a person to know if CBT is effective? Meta-analyses can help, but only so much. To paraphrase Eysenck: if you put garbage in, you get garbage out.

Even if we were dead-set on relying on the research, we’d be able to find evidence arguing for nearly every position: CBT is more effective than other psychotherapies, but only for certain conditions; it is not less effective than other psychotherapies (lol); it is no more effective than other psychotherapies; it is less effective than other psychotherapies; and it is statistically superior in ways that are neither robust nor clinically meaningful.

In fact, the only position I cannot seem to find supported by a meta-analysis is the one CBT’s reputation would lead you to expect: that it is clearly better than other psychotherapies in general.1

Usually, I am a fan of data over anecdote. But when the data are this unreliable, I fall back on experience and judgment. I also think that, at a certain point, too much emphasis on the head can dull the stomach. We can become so preoccupied with what can be proven that we forget what can be sensed.

A more straightforward way to judge CBT is to watch it work. What kind of conversation does it produce? What does it focus on, or hurry past? What sort of person does it imagine the client to be? What kind of therapist does it train someone to become?

If we approach CBT this way—not as a stack of studies to be litigated, but as a human exchange to be observed—I think it becomes obvious what it is.

It’s lame.

A Model Session

So, what does CBT look like in practice?

The following example comes from Chapter 3 of Healing by Tom Insel, former director of the National Institute of Mental Health. Insel presents this case favorably—as an example of the kind of care our mental health system could, but too often fails, to deliver.

The client, Sophia, seems to be depressed. She is struggling to meet basic responsibilities and has recently become self-critical. She begins working with a CBT therapist, Dr. Chou.

In the first session with Sophia, Dr. Chou asked her specifically about problem areas and focused on ways Sophia perceived the world, what Dr. Chou called ‘negative thinking.’ For instance, Sophia mentioned that she had failed to get the girls to day care on time twice in the previous week. Dr. Chou asked her about the verb ‘failed’ and questioned whether she could imagine delaying taking the girls to day care as a chance to spend more time with them. She also challenged Sophia’s sense of being ‘useless’ because she was on the sidelines during the Black Lives Matter demonstrations. Dr. Chou described the importance of looking at this pattern of negative thinking, loaded with self-judgment and blame. She gave Sophia a homework assignment to track these kinds of thoughts in a daily journal. Her task each day was to challenge this tendency to see herself only as a failure, unable to meet her own unrealistic expectations.

Is it just me, or is there something pathetic about this exchange? Many words come to mind—corporate, medicalized, dismissive—but none captures it as cleanly as this: if I knew nothing about Sophia, I would assume she was a child. Because I would only talk to a child like that.

Notice what happens when Sophia describes “failing” to get her children to daycare on time. Dr. Chou doesn’t explore why it happened, whether it matters, or what it means in the context of Sophia’s life. Instead, she moves quickly to soften the language—treating the word failed like a stain to be wiped off the counter—and then offers a more palatable interpretation that, personally, I would find downright insulting: that being late is an “opportunity” for Sophia to spend more time with her children.

Is Sophia speaking to a motivational poster or a person?

The session proceeds in a familiar direction: identify the unhelpful patterns of thinking, track them, challenge them, replace them. The subtext is: “Your interpretations don’t seem to be helping you. Try more useful ones instead.”

Not once does Dr. Chou seriously engage Sophia’s underlying concerns. The interaction stays on the surface, as if polishing a few words and reframing a few thoughts can lift the fog from someone’s soul. Sophia’s session amounts to a pat on the head and a “there, there, everything will be all right.” All that’s missing is the lollipop on the way out.

If I were Sophia, I might leave the appointment feeling more alone than before. Is this what passes for two people speaking authentically about the dilemma of being alive?

CBT’s tendency to reframe rather than understand is a problem because, in therapy, being understood is not a warm-up act before the real intervention. Sometimes it is the intervention. Often, as Camus says, “we merely wish to be pitied and encouraged in the course we have chosen.” We wish to be seen and sat with; to take a breath and look at our lives with someone else; to tell our story to someone who isn’t too busy, or too overwhelmed, to hear it.

I know this may not sound like much, but trust me: it is often enough to make the difference. The client leaves a little lighter, a little less tangled in whatever had been holding them down. That, really, is the magic of therapy: the thing so hard to explain. But when a CBT therapist rushes to correct a client’s thinking, the client can understandably feel as if they have been handled, and not heard.

CBT’s reframings are a problem for a second reason. As discussed in Part I, they often teach clients to look away from important information. Sophia’s self-criticism, for example, isn’t something she is conjuring out of thin air. It is tied to real pressures: her responsibilities, her children’s outcomes, her reputation. Discomfort is the cost of taking these pressures seriously—a discomfort Dr. Chou seems unable, or unwilling, to tolerate.

Part of the reason CBT’s reframings don’t receive more criticism is that people are unwilling to admit how judgmental they really are. Fortunately, I suffer from no such inhibition, especially when the reward is knocking CBT down a peg. So here’s what went through my mind as I read the vignette:

Think me too judgmental? That’s a judgment.

But unless I am uniquely mean, my first reactions show why self-criticism exists. Humans are judgmental creatures. If you don’t judge yourself, others will happily do it for you.

The point of self-judgment, of course, is not to suffer forever. It is to notice what needs to change—and act. If Dr. Chou understood this, she might have asked Sophia what her self-criticism was trying to tell her, and what was preventing her from acting on it. Instead, Dr. Chou tries to fix the feeling rather than listen to its message.2

At best, this kind of intervention offers temporary reassurance. A professional voice, standing in for the broader community’s judgment, tells the client, in effect: “You’re okay. You can relax.” But that is not the same as healing. Even if Sophia had felt relief after her work with Dr. Chou, it would not necessarily mean she had discovered a better way to think. She may have simply been granted a social permission loan.

Unfortunately, Sophia experienced no such relief. “For Sophia,” Insel writes, “the weekly visits, the homework, and the expectations of improvement proved too much. She discontinued CBT after four sessions.” Yes, it sure sounds like the weekly sessions were the problem.

I was surprised to see this from Tom Insel, whose book I otherwise enjoyed. It made me wonder what kind of person, if any, would be drawn into psychotherapy by an example like this. The exchange between Sophia and Dr. Chou reads less like an authentic conversation about complex human problems than a scripted exercise in therapeutic hygiene. CBT tries to sweep emotion away before asking what it knows.

If therapy is headed in this direction, it should be handed over to AI.

The CBT Type

Of course, one can always object that this is only one example. Then again, it was a favorable example—CBT in its best light—and it was still lame as hell.

But we don’t need to rely on one example. As a practitioner in the field, I’ve had plenty of run-ins with CBT—anecdotal experiences which, under the circumstances, I hope you’ll allow me to admit as evidence. For background, I got my master’s at La Salle, which had more or less adopted CBT as its main orientation, and did my internship at a psychoanalytic institute. So I got to see Godzilla versus King Kong up close.

Wanting to be an even-handed critic, despite the lopsidedness of my experience, I sometimes wonder how much of my disdain for CBT comes from the introductory class I took on it in graduate school. Among the many boneheaded things my teacher suggested, one stands out. “At the end of the first session,” she advised, “ask the client how much they trust you on a scale from 1 to 10. If it’s not already a 10, ask what you can do to get there by the end of next session.”

I’ve remembered this advice for over a decade, mostly because it is hard to imagine a more revealing misunderstanding of human nature. A client who reported complete trust in me after less than two hours would alarm me far more than one who did not. Trust must be built—I can’t believe I’m having to type this—slowly, over time.

This is what I mean by “hygienic.” CBT all too often approaches the messiness of life with a spray bottle and cloth: we’ll get that taken care of for you right away. Don’t trust me yet? What could possibly be getting in the way? Think you’re no good? Where’s the evidence? Keep ruminating about the future? Stop it.

Clients are not being met as much as moved along: toward cleaner thinking, better habits, and improved outcomes. And many of them can feel it.

Indeed, the most common complaint I hear from clients who have tried and disliked CBT is that it felt disrespectful toward them and their problems. It treated them as if they were stupid, and their problems as straightforward and easily fixable. They experienced it as patronizing, procedural, shallow, and obvious.

Granted, I advertise myself as someone who “works best with those who think they are too smart for therapy,” so my caseload may be particularly sensitive to being patronized. But still.

A few clients have gone further, saying the method itself seemed to prevent them from forming a genuine connection with their therapist. In CBT, you see, conversation is kept on topic. The number of sessions is often capped. Progress, or the lack of it, becomes the organizing concern. Hell, the treatment is often manualized, the point of which is to make therapy less dependent on the particular person doing it.

Isn’t this odd in a field where the relationship itself is supposed to matter so much?

I do not mean to paint every CBT therapist with the same brush. There are gifted therapists and hopeless hacks in every orientation. But orientations do not randomly attract people, and CBT offers a very particular role: the active, structured, correcting, scientifically-minded guide. That role can bring real virtues—focus, practicality, accountability—but also characteristic vices: impatience, hubris, and a faintly managerial attitude toward suffering.

Nor does a method merely attract a type. Over time, it can cultivate one. Another writer whose work I otherwise respect, Abigail Shrier, says that “a good therapist should do what cognitive behavioral therapists do: prove to a patient that rumination is an unproductive mode of thought and train them to stop.”3

Are we talking about dogs or people here? And what kind of therapist does one become after twenty years of this?

In the end, perhaps the most damning thing I can say about CBT is that it’s entirely possible, in a CBT session, for the therapist to do most of the talking.

Doesn’t that just seem—wrong?

Why Does CBT Exist?

Believe it or not, I don’t think CBT is entirely useless. I can sympathize with it on three fronts.

First, CBT’s basic frameworks can be genuinely clarifying. What sounds insultingly obvious to one client may be helpful to another. I might find CBT’s core premise—that thoughts, emotions, and behaviors are interconnected—too obvious to belabor, but I work with this stuff every day. Not everyone does.

Second, CBT is a response to legitimate criticisms of psychotherapy. Therapy has long been expensive, opaque, and difficult to evaluate. If you want to provide mental health care to more people at lower cost, you’ll naturally be tempted to isolate the active ingredients, simplify the method, manualize the treatment, measure the outcomes, and make the whole thing portable. Some kind of McTherapy will likely be the result.

Finally, and annoyingly, CBT does get some things right. Identifying the belief beneath a feeling can be useful. If someone cuts me off in traffic, my anger is usually supported by some assumption: that people should not drive this way, that I have been disrespected, that I am being taken advantage of. Even if I do not abandon the belief, it helps to see that my anger depends on it. More generally, CBT can help people interrupt habits, test beliefs, and find something solid to do when they are drowning in vagueness. Structure, all by itself, can help.

But this is also the problem: CBT is too pleased with its clever little programme. In identifying a few true and useful things, it tends to miss the bigger picture: therapy is more about process—a big part of which is the relationship—than content. As Ivan Illich warned back in the 1970s, the modern doctor “prides himself on the knowledge of pain mechanics and thus escapes the patient’s invitation to compassion.”4

In short, clients deserve better from the best my field has to offer.

Why Does CBT Persist?

If CBT is so lame, why is it still here?

In Walden, published in 1854, Thoreau observed: “I perceive that we inhabitants of New England live this mean life that we do because our vision does not penetrate the surface of things. We think that is which appears to be.”

That is precisely the problem with the session between Sophia and Dr. Chou: it appears helpful. It seems legitimate. An insurance company would trip over itself trying to reimburse it. Indeed, the reason I settled on the word “lame” to describe CBT is that there isn’t a more obviously damning adjective. On the surface, at least, it has all the trappings of scientific authority and professional credibility you could want: it is evidence-based, measurable, expert-led, and rational. What’s the issue?

The issue is that these are just words. What matters is whether there is any substance behind them.

For example, in Where Is the Evidence for “Evidence-Based” Therapy?, Jonathan Shedler argues that the term “evidence-based,” in the context of psychotherapy, may be “a perversion of every founding principle of evidence-based medicine.” He claims the evidence actually shows that such therapies “are ineffective for most patients most of the time.”

Measurement has a related problem: it is only as helpful as the measure is valid. The ability to measure something tempts us into thinking we are measuring what counts. This is how a client can get through ten CBT sessions and still feel there is plenty of work left, only to be told that, according to their Beck Depression Inventory scores, they are much better now and ready to be discharged.

Expertise, too, is not always as foolproof as it seems. As my favorite therapy writer, Kenneth Fisher, puts it: “No doubt, the therapist is a specialist; still, considering what he is faced with—the complexity of life itself—can there really be any expert?”5

Then there is rationality itself. In Part I, I argued that CBT, despite its emphasis on rationality, reasons quite poorly about it. But the deeper problem is that reason is not the whole of wisdom. Even in Western thought, there is plenty of recognition of this—that “the height of reason,” as Fisher says, “is to know when it no longer serves, when some other response is called for.”6 Josiah Royce makes a similar point in Race Questions, Provincialism, and Other American Problems (1908): “As a professional reasoner, I have a profound contempt for deliberate excesses in the work of reasoning.”7

Evidence, measurement, expertise, rationality. None of these are bad things. That is what makes them dangerous. CBT reminds me of that student—God bless him or her—whose essay satisfies every criterion on the rubric, forcing the teacher to come up with another way to say: “Yeah, but it still sucks.”

Likewise, CBT hits every point on our culture’s rubric. It masquerades as the hero we asked for. It belongs to a society increasingly suspicious of whatever cannot be reduced, explained, proven, scaled, operationalized, measured, or, in many cases, monetized. The result is not the old nonsense psychotherapy was once awash in, but something much harder to criticize, much harder to dislodge: a professionally respectable way of missing the point.

Of course I worry sometimes that I’ve gone too far—that I’m waterboarding a perfectly decent approach that helps plenty of people make better sense of their thoughts and emotions. But then, invariably, I return to a novel, play, poem, or short story, and am reminded that if I could teach CBT one thing, it wouldn’t be the evolutionary perspective. It would be the depth found in the humanities. We’ll turn there in Part III.

1. Claims to be rooted in reason, science, and evidence, but misunderstands human reasoning, neglects the best available science, and overstates the evidence for its effectiveness (Part I).

2. Can be superficial in ways that diminish or overlook client suffering (Part II).

3. Crowds out richer ways of understanding and addressing distress (Part III).

There are dozens of therapeutic orientations—over seventy currently listed on Psychology Today—and many of them are pseudoscience. So why spend three posts critiquing just one of them? Because while there may be seventy-plus approaches in theory, functionally there is increasingly only one. Like an invasive weed, CBT is taking over the field.

CBT may not have won the game of discovering the best therapy, but it has undeniably won the game of convincing everyone it has. It dominates training programs, treatment guidelines, research funding, and public perception. Hell, my guess is that the average reader can name two kinds of psychotherapy: CBT, which they would identify as the gold standard, and psychoanalysis, which they would regard with a healthy dose of suspicion.

Convinced of its supremacy—or at least unaware of alternatives—the average person thinks: “CBT or bust.” And that’s a problem, because CBT is not what it has so successfully advertised itself to be.

False Advertising #1: CBT is the therapy of reason.

Here is an AI summary of CBT, with my emphasis italicized:

CBT is a relatively modern form of psychotherapy, developed in the 1960s and 70s by figures like Aaron Beck and Albert Ellis. At its core is a simple idea: that our thoughts, feelings, and behaviors are tightly linked, and that psychological distress is often maintained by distorted or unhelpful patterns of thinking. By identifying and correcting these patterns—challenging irrational beliefs, testing assumptions, and replacing them with more accurate or useful ones—CBT aims to reduce symptoms and improve functioning. It presents itself as a structured, goal-oriented, and above all scientific approach to therapy, one that prioritizes evidence, measurement, and rational analysis.

Oh boy. Where to begin?

Let’s start with CBT’s “core belief,” if you will. CBT is organized around the belief that much of human suffering is caused or maintained by faulty patterns of thought and can therefore be relieved by better ones. If that premise is wrong, then CBT’s approach is wrong.

That premise is wrong.

It’s wrong because it proceeds from an incorrect assumption about what these patterns are and why they exist. As much as CBT might wish otherwise, human beings are not formally or academically rational. We are ecologically rational instead. Our minds evolved in ancestral environments where decisions had to be made quickly, under pressure, and with incomplete information. That means relying on “fast and frugal” mental shortcuts—heuristics—that are often more adaptive than accurate.

Just because an evolved heuristic does not always produce the response a therapist—or even a client—would prefer does not mean it is “irrational,” “distorted,” or, God forbid, “maladaptive.” Often, the issue is not the heuristic itself, but the task, environment, or framework being used to judge it.

Now, a defender of CBT might reply: “Fine, some heuristics are useful, but others are exaggerated, misapplied, or harmful in the client’s present circumstances.” Fair enough. But that concession changes the whole frame. The question is no longer whether the client is thinking rationally or irrationally, but whether an otherwise useful system is calibrated to its current context. That is a very different posture from identifying “distortions” and correcting them.

Without understanding why certain patterns of thinking, from overgeneralization to catastrophizing, might have evolved, CBT has no reliable way to locate the source of a client’s problem. Is the client distressed because their situation is genuinely bad? Because previous experience has made them more sensitive to this particular threat? Because a once-useful heuristic is being activated in a mismatched modern context? Because the mechanism itself is impaired?

Without asking what these patterns are for, CBT falls prey to Chesterton’s fence: trying to remove something without first understanding why it was put there. Most of what CBT calls distorted, illogical, or irrational has a perfectly intelligible evolutionary rationale. This is indeed what the phrase “cognitive distortion” betrays: a failure to distinguish design from dysfunction.

For example, take a look at the list of cognitive distortions below, variations of which can be found on CBT websites and training materials:

Cognitive distortions like these are CBT’s basic unit of analysis: the “errors” it is trained to detect and correct. Much of CBT’s value comes from helping clients identify these patterns of thought and replace them with better ones.

Let’s focus on two: mental filtering, or focusing more on negatives than positives, and mind reading, or assuming you know what others are thinking.

A client does not need a “negative script,” a gloomy childhood, a chemical imbalance, or low self-esteem to over-attend to negatives. They only need to be human. People over-attend to negative information because it often carries more informational weight. “Susie was offended by something you said” calls for a response. “Susie seems to like working with you” doesn’t.

As for mind reading, might there be a reason humans routinely hypothesize about what is happening in other people’s heads—especially when it concerns them, and especially when it might be negative? Might such a running calculation, uh, help us maintain a good reputation, which is pretty important for one of the most social species on earth?

To be fair, CBT is not usually warning against social inference as such. It is warning against treating guesses as facts. Many CBT sources label mind reading a distortion because it involves making assumptions without evidence. But when do we ever have full evidence? People do not typically announce their negative thoughts about us. Social life requires us to form theories with limited information. While these theories might not always be accurate, that doesn’t make the habit of forming them maladaptive.

This is the recurring problem: CBT sees the inaccuracy but often misses the function. It pronounces defects before understanding design. For a therapy whose central promise is to improve how clients think, that is no small problem. In answer to one of the client’s most basic questions—“Why do I think this way?”—CBT’s answer all too often seems to be: “Don’t worry about that. Just think this other way instead.”

False Advertising #2: CBT is the therapy of science.

Ok, someone might say. Let’s call these patterns of thought heuristics rather than distortions. Problem solved?

Not quite. Once the question shifts from “How is this thought distorted?” to “Is this heuristic useful in this context?”, we have to think more carefully about what “useful” means. This is where several findings from the scientific literature become relevant:

1. Bad feelings are often functional.

2. Evolved systems track cost, not probability alone.

3. Many heuristics remain remarkably effective, even in modern contexts.

In other words, a thought pattern can be painful but functional, inaccurate but protective, simple but effective.

These are not obscure findings, by the way. They are hiding in plain sight—in the science CBT so often invokes.

Bad feelings are often functional.

Just because something makes us feel bad doesn’t mean it’s malfunctioning. Often, quite the opposite. One of evolution’s more common tradeoffs is bad feelings for adaptive outcomes. For example, there aren’t many feelings worse than a deep and searing shame. Then again, shame motivates us to make amends, repair our standing, and avoid similar actions in the future.

A therapist who tries to remove shame without understanding the role it is playing risks treating a fever as the disease.1

Evolved systems track cost, not probability alone.

If Possibility A is no more likely than Possibility B, but much more dangerous if true, an adaptive system should lower its detection threshold for A. “Better safe than sorry” is not a cognitive error. It’s a design principle.

It is also the basic logic of signal detection theory, which has been available to psychologists for decades.

For example, have a look at the image below. What do you see?

If you saw a snake, what did it cost you? A brief flicker of fear, probably gone by now. But if you saw a stick, and it was actually a snake—and could somehow jump through the computer—you might be dead before the end of this article. That’s why the mind would rather mistake a stick for a snake than a snake for a stick.

Isn’t personalization, or “making everything about oneself,” a social version of this? A client might be overestimating the probability that someone else’s behavior is about them, but not necessarily the cost. The cost of a false positive is some distress; the cost of missing that someone is annoyed, offended, disappointed, or pulling away may be a lost chance to repair the damage.

A therapist who focuses on probability alone, rather than probability × cost, may not be correcting distortion so much as disabling judgment.

Many heuristics remain remarkably effective, even in modern contexts.

Evolved heuristics are not necessarily crude leftovers from a vanished world. Many outperform more complex, deliberative models built with far more information and computational power. Gerd Gigerenzer and colleagues, for example, have shown that the recognition heuristic can perform extraordinarily well: subjects with less knowledge of a domain, such as tennis, were better able to predict tournament outcomes than those with more knowledge, often by favoring the more recognizable player.

A therapist who tries to replace one of these heuristics should remember that it has been tested over a very long time.

The points above are not speculative. They are findings from psychology and related sciences that have been available for decades. If CBT is going to position itself as the most evidence-based, scientifically grounded approach, it should at least avail itself of, you know, the scientific literature. By and large, it hasn’t.

If anything, CBT seems to be moving in the wrong direction. Over the past fifty years, it has gone from calling these patterns of thought illogical, to irrational, and most recently, maladaptive. So much for incorporating the evolutionary perspective.

Let’s bring the last two sections to a head. What would a reasonable, scientifically literate person conclude about what CBT has called “cognitive distortions”? That they are better understood as heuristics—fast, frugal, context-sensitive shortcuts—which means they are often:

1. Not unique to the client in question.

2. Calibrated to expected cost, not mere probability.

3. More effective than they are given credit for.

4. Hard to replace with something better.

Taken together, doesn’t this cast serious doubt on CBT’s central project of “correcting” them?

False Advertising #3: CBT is the therapy of evidence.

At this point, the CBT advocate, heavily under siege, pulls out their strongest defense: CBT’s track record of success. “To hell with your academic arguments,” they say. “The proof is in the pudding. CBT works better than pretty much everything else.”

My reply? It doesn’t.

Across meta-analyses, differences between bona fide therapies are often small, inconsistent, or disappear over time—hardly the decisive advantage CBT is often said to have. In many cases, CBT performs roughly on par with other structured therapies, suggesting that its specific theory of change may be doing less work than advertised.

The main way CBT has consistently and indisputably outshined its peers is in the number of studies it has run—in the number of times it has beaten its chest and shouted from the rooftops that it’s the best. Unfortunately, as Leichsenring and Steinert point out, “more studies do not necessarily imply that a treatment is more efficacious.”

Keep in mind: CBT is effective. It’s just not more effective than other psychotherapies most of the time.

However, let’s say for argument’s sake that it is. Even then, its success would not necessarily prove that its theory of change is correct. It could instead be an unusually compelling placebo. Two reasons stand out. First, conventional wisdom says that CBT is the best, and placebos tend to benefit from shared belief. Second, CBT is active. Its practitioners tend to be confident, focused, and generally scientific-seeming. They have plenty of jargon and credentials. Odd as it may sound, the mere appearance of rigor, knowledge, and competence can be helpful even when the theory behind them is wrong.

So, to me, the conclusion is hard to avoid. Rather than calling CBT the most evidence-based, we should be calling it the most effectively promoted. I’m more inclined to hire a CBT therapist for my marketing department than for my mental health.

End of Part I

CBT’s core promise is that much suffering can be relieved by improving the way people think. But this promise rests on a flawed understanding of thought itself. In many cases, we are not built to reason the way CBT wants us to.

In other cases, we shouldn’t. CBT does not fully appreciate that bad feelings can be functional, that evolved systems track expected cost rather than mere probability, and that many heuristics perform remarkably well even in modern contexts. What CBT calls distortion may often be ecological rationality viewed from the wrong angle. All too often, CBT is trying to improve or fix what it does not understand.

Finally, when all else fails, CBT appeals to its strongest refrain: “Evidence is key! CBT is the most evidence-based psychotherapy available!” Under closer inspection, however, this claim is not supported by the evidence.

To be fair, CBT deserves some credit. Not only did it stick its finger in the wind and divine the direction the field was headed, it helped define that direction. In large part, we owe it to CBT that psychotherapy is now so obsessed with science, evidence, measurement, and the concrete question of what works. Much of the field beforehand was a wasteland of speculation—a bunch of “anything goes”—dominated by charismatic leaders and compelling theories.

CBT pulled the conversation forward, but in the process, it overstated its understanding of the mind, its command of the science, and its evidence of superiority. Psychotherapy remains a complex, noisy domain, and we are still far from clear, decisive answers about what works best.

My sense is that the main remaining counterargument from the ranks of CBT will be some version of: Have you looked at this recent meta-analysis? What about the work of so-and-so? Have you seen the results for phobias? Eating disorders? Past a certain point, I confess to finding this evidence-swapping tiresome and unproductive. In the next article, I’ll argue that we would do well to heed the words of Potter Stewart, the Supreme Court justice who, when asked to define obscenity, said: “I know it when I see it.”

In a similar way, I think we can just look at a typical CBT session and call it for what it is: it’s lame.

First, bottlenose dolphins don’t trade. As far as I know, there is no case in which one dolphin gave another dolphin a mackerel in exchange for some sardines. Also, as far as I know, there is no record of a dolphin who gave another dolphin a mackerel today expecting to get a mackerel back tomorrow. These creatures don’t have, as humans do, the “propensity to truck, barter, and exchange.”

Second, bottlenose dolphins do form alliances. The males of this species will cooperate with one another, repeatedly, to aid one another in sequestering females for mating.

I think these two observations are important because they tell us why dolphins have friends. They don’t have friends so that they can reap gains in trade. Their friends aren’t exchange partners. Their friends are allies. Dolphins use other dolphins to increase their chances of mating.

The available evidence makes the answer to the question of what dolphins friends are for pretty clear.

In contrast, observations in vampire bats (D. rotundus) raised the possibility that they were using friends as exchange partners. Early work suggested that a bat who was successful in the hunt today gave blood to another bat, who would return the favor when they were successful in the hunt. Vampire bat exchange remains an area of some controversy, but for my purpose, the key point is that the long-term relationships among bats—friendships—are, putatively, for exchange but not for alliances. Assuming that this proposal holds up, the answer to the question of what bat friends are for seems pretty clear.

In humans, there is still disagreement about what friends are for. Alliances? Trade? Something else?

Early in my career, many wrote about this question as if it were already solved. There were two reasons people had friends. One was that humans help relatives (because of kin selection) but sometimes we treat friends as family. The second reason drew on the logic of reciprocal altruism developed by Robert Trivers, who suggested that humans play Tit-for-Tat, responding to cooperation with cooperation, as in the bat case.

In 2003, anthropologist Joan Silk, writing about the literature in evolutionary approaches to human behavior, noted that “[w]hen friendship is mentioned, it is usually assumed to be the product of kin selection, which is misdirected toward nonkin or Tit-for-Tat reciprocity” (p. 46, my italics for emphasis).

John Tooby and Leda Cosmides proposed an alternative to the conventional wisdom, using the “banker’s paradox” to illustrate their idea. If you are a bank, you want to lend money to those most likely to repay it. The “paradox” is that the people in the best financial shape are least likely to be in need of money. Tooby and Cosmides argued that a key problem humans faced over evolutionary time is that we sometimes fall on hard times. How do we convince others to lend us a hand? The key is to have made yourself irreplaceable, to have some skill or resource that your friend cannot easily do without. Friends are, according to this view, insurance, helping you to buffer against the risk of a sharp downturn in circumstances.

So there are at least three possibilities. Friends might be exchange partners, as early scholarship mostly assumed. Friends might be allies, as with our bottlenosed friends. Or friends might be insurance policies, a la the banker’s paradox.

And, of course, friends might be a mix of these and more.

My former student, Peter DeScioli and I, were surprised to find that the allies hypothesis was not widely discussed. (Peter included this idea in his doctoral dissertation.) Our sense is that Joan Silk’s remark was just right: whenever people talked about friendship, they talked about reciprocity.

Generally, we thought there were some problems with the idea that friends are exchange partners. I’ll just give two.

First, people certainly don’t seem to behave as if they are. If I have you over for dinner and you pay me $100, our friendship will be damaged, not improved. When it comes to friends, we humans make a big deal about the idea that it’s not an explicit exchange relationship. We eventually get irritated if a friend doesn’t reciprocate, but we also don’t keep close score. It is “cooperation without counting.”

Second, people get jealous about friendships in a way that echoes jealousy in romantic relationships. It’s fairly clear at this point why romantic jealousy exists. Other people can threaten one’s certainty of paternity or the investment one is getting from a mate. If friends are trade partners, then why are others a threat? In other contexts, it’s actually better for someone if their exchange partner has other trading relationships. Those ties provide more opportunities for gains in trade. See Jamie Krems’ work on friendship jealously for more.

None of this is to say that humans don’t trade. Of course they do. But trade relationships are different from friendship.

Suppose that human friendships are not, as many assume, like bat relationships—reciprocal exchange—but are, as I believe, like dolphin friendships—used for alliances. Further suppose a conflict emerges and your friend has to pick a side. If they are (just) a trading partner, you might not be that irritated if they side against you. The U.S. trades a lot with China, but doesn’t expect that country to take our side when we fight with, say, Venezuela or Cuba. But if friends are allies, then you will probably be irritated if they side against you when a fight starts. After all, the whole point of allies is that they side with you.

That idea brings me back to the topic of a prior post, about loyalty, and the results of a new paper just published in the journal Evolution and Human Behavior. Landers and colleagues—I’ll use LDMS to refer to the authors—were interested in the tension I discussed, loyalty versus morality. They begin their research report with a simple question: “How do we judge someone who must decide between loyalty to their friends and adherence to moral principles?”

Notice, first, that implicit in this framing is that these two things—loyalty and morality—are ways to choose sides. (Loyalty is not, as some would have it, a kind of morality.) Their main prediction has to do with this view. If morality is for side-taking, then there might be one particular circumstance in which someone excuses their friend (and putative ally) for siding against them: when they have committed a moral wrong.

LDMS conducted a series of studies, and this pattern is exactly what they found. Suppose John takes credit for an idea at work, but he actually got the idea from Casey. When John tells his friend David about this, David says, “I think you were in the wrong. I’m with Casey on this one.”

When subjects were given this vignette, they rated David as less disloyal compared to control cases in which John hadn’t done anything wrong. That is, you’re seen as less disloyal if you choose sides against your friend when your friend did something wrong.

If that’s confusing, here is the basic idea, drawing on the time that the Unabomber’s own brother turned him in:

Humans expect their friends to join their side when disputes arise because friendships are alliances. So when someone isn’t loyal to a friendship, they need to justify it. This justification is often to do with morality, the other way humans choose who to back.

In some ways, it’s weird that there is still debate about what friends are for and what morality is for. For dolphins and bats, the explanations aren’t necessarily universally accepted, but the debate is narrower. Shouldn’t the social sciences have made more progress on this question for humans?

I don’t really know why they haven’t, but here are some guesses.

First, to be fair, humans are complicated. So figuring people out is more difficult than figuring dolphins. And probably the answer will be a mix of different explanations, with some other ingredients thrown in.

Second, it’s easier to think about trade, a two-player game, than about alliances, which is a game that requires more than two players. Multi-player games are just harder to think about, so social scientists gravitate toward the easier, two-player games.

Related, there’s something of a primacy effect. Trivers’ work in the 70s was paradigm-changing. He offered one of the earliest accounts of the evolution of cooperation and science has a great deal of momentum. To return once again to Joan Silk’s remark, Trivers and reciprocal altruism became the default explanation for cooperation, and that has carried through for decades.

So, are your friends more like bats, ready to give you a blood meal when you’re hungry, or more like dolphin buddies, ready to stand next to you when you need an ally?

My sense is that while of course a friend will give you some food when you’re in need, their real value lies far beyond exchange of goods and services over time. The work by LDMS, to my eye, adds evidence in favor of this view.

Before concluding, I think it’s important to add a somewhat subtle point.

You might be tempted to think that when someone betrays a friend on moral grounds, that indicates a certain kind of integrity or virtue on that person’s part. Surely we want people to stand on principle, even if it means siding against a friend.

I would resist that temptation.

The excellent film A Man for All Seasons tells the story of Sir Thomas More and his eventual execution on the order of Henry VIII. More refuses to take the Oath of Supremacy, endorsing the king’s assertion to being the Supreme Head of the Church of England. He refused on the grounds that Jesus Christ had established that the Pope, not a secular ruler, was the head of the church. More’s refusal to take the oath puts him at odds with the king’s desire to annul his marriage to Catherine of Aragon and marry Anne Boleyn.

More is careful. He never actually says why he won’t take the oath, and he argues that because he is simply refusing to act, rather than taking an action, he can’t be prosecuted for treason, taking a stand against the sovereign. (More was an early proponent of the act/omission distinction.)

In the movie—which tracks the real history reasonably well—More’s younger protégé Richard Rich has an opportunity. If he betrays More, claiming that More did, in fact, deny that the king was the true sovereign, then he can gain the king’s favor, allowing the king to condemn More to death. Rich does exactly this, claiming, falsely, that More made the treasonous remarks in a private meeting. Rich gets a fancy title as a result—Attorney-General for Wales— leading to More’s famous line, “Why Richard, it profits a man nothing to give his soul for the whole world... but for Wales?”

Siding against a friend who did wrong might be the tough-but-heroic thing. Certainly, it’s easy to respect David Kaczynski for his role in catching his brother.

But this misses the subtle and vexing aspect of morality. In the case of More’s remarks, there was no evidence for Rich’s claim other than Rich himself. This circumstance is frequently the case regarding moral disputes, when evidence is scarce or invisible. This fact means that people can often easily assert moral wrongdoing or go along with others’ assertions of their friend’s wrongdoing. If one claims to believe that the Jews are responsible for the country’s woes, for example, it’s easier to watch as your Jewish neighbors are collected by the SS. A key feature of the work by LDMS is that people excuse disloyalty when they judge their friend to have been immoral even if the friend themselves didn’t think that they did something wrong. People use their own beliefs about wrongness when excusing themselves from being disloyal.

So, claiming to believe that a friend did something wrong can be a powerful way to act in one’s own interest—pleasing the king, avoiding the gaze of the gestapo—and, in addition, avoid the cost of being seen as disloyal.

Not every case of disloyalty in the name of morality is like the Kaczynski case. As with all human behavior, seeming virtue is often driven by darker motives.

The Argument, in Brief

• User manuals catalog the characteristic ways a product can break down. These breakdowns are not random; they follow from how the product is designed.

• The same is true of the mind. Its mechanisms create its characteristic routes to dysfunction: simulation can become rumination, attachment can become grief, orientation can become disorientation.

• Designed systems tend to break down in three broad ways: malfunction, deprivation, and misuse. Each requires a different kind of repair.

• The DSM is best understood, for now, as an error log. But that’s not nothing. These errors can help us reverse-engineer how the mind was built.

Users Manual

Not too long ago, my wife and I bought a house, which resulted, among other things, in my reading a bunch of user manuals—for smoke detectors, water heaters, washing machines, and the other machinery of domestic adulthood. You can imagine how discouraging this was to someone who has long spent his morning hours with “the noblest recorded thoughts of mankind,” but I consoled myself that the change was temporary, and it was. Plus, it taught me something about the human mind.

A common takeaway from user manuals is that there are more ways for things to go wrong than right. An oven can burn us, poison us, crush us, or blow our house up. It can get dirty, lose a part, or disconnect from a power source. In the best-case scenario, it simply cooks our food.

But these failures are not random or infinite. An oven cannot give us a gambling addiction or sprain our ankle. Its failures are the shadow of its success—an inescapable part of its design.

To understand how something is designed, one must ask two basic questions:

1. What is the thing trying to accomplish? This is its function.

2. How is it trying to accomplish it? These are its mechanisms.

The function of a washing machine, for example, is to wash clothes—aren’t you glad you’ve subscribed to this brilliant Substack?—and its main mechanism is agitation, powered by electricity and aided by soap and water.

Already, the possible failures begin to suggest themselves. Because it uses water, a washing machine can leak, flood, rust, or grow mold. Because it uses electricity, it can short-circuit or catch fire. Because it relies on agitation, it can stretch, tear, or mangle clothing.

The function of a thing tells us what it is trying to do, and its mechanisms tell us how it tries to do it—which also suggests how it might fail.

Meanwhile, if you know what a thing is trying to do, you can often make a decent guess about the mechanisms it will use. If something is meant to wash, it is a good bet that water will be involved because water happens to be very good at loosening, dissolving, carrying, and rinsing dirt away. Water is, as Daniel Dennett would say, a “good trick” for cleaning, just as the wheel is a good trick for transporting or unlimited vacation policies for keeping people at work.

In biology, when the same trick appears in unrelated species, it is called convergent evolution. The wings of bats, birds, and insects evolved independently because flight is a useful way to exploit a niche: to escape predators, reach food, find mates, and occupy places otherwise difficult to reach. But the same wings that make flight possible also bring falls, injuries, failed takeoffs, and collisions into the picture.

Every mechanism opens a path through the world—and every path has its edges.

The Manual of Mental Disorders

Generally, the more something tries to do, the more complex it has to be. It used to be that refrigerators only cooled food. Now they make ice, filter and dispense water, connect to the internet, and berate you when you’re out of milk. Each of these new functions creates a new point of failure. There’s a reason the manual for my washing machine is longer than the one for my daughter’s night-light.

If user manuals get larger with complexity, what should we expect from the “manual” for the most complex thing we know of? Yes, I’m speaking of the Diagnostic and Statistical Manual of Mental Disorders, a title that hopefully sounds less strange than it did a few minutes ago.

Now, the DSM isn’t strictly a user manual because it contains only one of the usual sections: the troubleshooting section, which tells us what can go wrong and how we can try to fix it (usually some combination of therapy and medication). It lacks the parts section, which would explain how the brain is pieced together, as well as the operating instructions, which would tell us the conditions under which the mind works best. In that sense, it is less a user manual than an error log. These errors have names like major depressive disorder, social anxiety disorder, obsessive-compulsive disorder, trichotillomania, anorexia nervosa, and post-traumatic stress disorder. For billing purposes, they are given convenient shorthands like F43.11.

The DSM is an enormous document for many reasons, but one of the more forgivable ones is that the human mind is trying to do an astonishing number of things. It has to perceive the world, interpret threat, regulate emotion, pursue status, attract mates, maintain friendships, raise children, remember the past, imagine the future, detect betrayal, infer intentions, respond to pain, follow norms, tell stories, worship gods, make jokes, and grieve the dead. As William James argued, humans are not exceptional because we lack the instincts found in other animals. We are exceptional because we have more of them.1

This robust functionality has obviously served us humans well. (It seems that every time I turn around, I see another one of us.) But it comes at the cost of more routes to failure. Because we can simulate, we can also ruminate. Attachment comes with the tradeoff of grief. The capacity for orientation means we can occasionally become disoriented. Without threat detection, there would be no anxiety.

The very capacities that allow the mind to function also create its characteristic routes to dysfunction. And for every function, we should expect not one possible dysfunction, but many. Success is a narrow path, after all; failure is an open field.

However, not all failures are of the same kind. User manuals are usually careful to distinguish between problems with the machine, problems with the user, and problems with the conditions under which the machine is being used. That is another way the DSM differs from an ordinary user manual: it does not reliably preserve these distinctions. Its disorders are mostly presented as if they belong to the same general class.

Three Ways Things Fail

Let’s say the touchscreen on your washing machine isn’t working. Explanations will fall into three broad classes. First, something inside the machine might have failed: the screen, the wiring, the control board. This is malfunction. Second, the machine itself might be fine, but something it needs is unavailable: the power is out, the outlet is dead, or the circuit breaker has tripped. This is deprivation. Third, you might be using it wrong—perhaps your fingers are too wet for the screen to register them. This is misuse.

At the level of function, these distinctions hardly matter. It’s just not working. This is why, when standing in front of a useless appliance, one does not typically say, “Ah, good, the touchscreen itself is intact, but one of its external dependencies has failed.”2 One says something less printable. But at the level of troubleshooting, these distinctions matter a great deal because they call for different solutions. If the touchscreen is broken, you call a technician. If the power is out, you check the breaker. If your fingers are wet, you dry your hands.

So, our three categories of failure are:

1. Malfunction: an internal part or process is not working.

2. Deprivation: a necessary input or condition is absent or unstable.

3. Misuse: the thing is being used outside its intended range.

Does the mind fail in the same three ways? It is no less designed than the washing machine, so I see no reason why it wouldn’t. But you’d never know this from reading the DSM.

Let’s take malfunction first. A person might suffer because some part of the mind, or some interaction among parts, is not doing what it is supposed to do. This is the category closest to the traditional disease model. For example, in Rob’s review of Sociopath, he argues that sociopathy may arise when certain social emotions—fear, guilt, shame, empathy, and compassion—fail to develop properly.

Then there is deprivation: failure caused not by something broken inside the mind, but by the absence or instability of something the mind needs. The mind relies on sleep, safety, belonging, status, movement, energy, sunlight, agency, meaningful work, and so on. As these supports disappear or become unstable, functioning tends to decline. For example, children raised without adequate exposure to human speech do not acquire it normally, even if the underlying machinery is otherwise intact.

Finally, there is misuse: failure caused by using the mind outside its intended range. Does this not describe the modern experience, with its infinite comparison, endless feeds, global moral emergencies, relentless task-switching, ultra-processed food, engineered stimulation, chronic sitting, and so on? Something optimized for one environment should be expected to behave strangely when run in another. This is the basic idea behind evolutionary mismatch, and you can find a useful discussion with examples here.

Are these not critical distinctions, pointing to radically different solutions? The DSM—and the field of clinical psychology more generally—doesn’t seem to think so. The industry appears content to call many different kinds of failure by the same name: mental illness.

To me, “mental illness” should be reserved for malfunction: conditions such as schizophrenia, intellectual disability, dementia, and some neurological or developmental disorders. Deprivations and misuses warrant different names. This is why I rarely use “mental illness” as a catch-all. Terms like “mental distress,” “misery,” or “suffering” are often more accurate, because in practice, I see all three kinds of breakdown.

Of course, the most common thing I see is not failure at all, but adaptive functioning with painful emotional byproducts. Has the mind really failed when it feels claustrophobic in an elevator? Depressed after a breakup? Socially anxious before public speaking? Miserable when isolated? Distracted by a device engineered to hijack attention? Disoriented because science has demolished so many former sources of meaning? Humiliated after being ranked against thousands of strangers?

As Randy Nesse would say, there are often good reasons for bad feelings. A bad feeling is not necessarily a broken one.

The DSM is good at describing dysfunction once it appears, but not so good at distinguishing where that dysfunction originates. This means it is also not especially good at prescribing solutions. Its go-to treatment recommendation of therapy, medication, or both is the equivalent of a user manual saying “call technician” for every conceivable failure of a washing machine. A better method would be to troubleshoot the source of the problem and proceed from there. Otherwise you might pay $100 for a technician to hand you a towel.

The DSM has a hard time troubleshooting problems because it lacks the operating-instructions section of a typical user manual: an account of what the mind was built to do, the conditions under which it usually works, and what kinds of demands fall outside its intended range. To be fair, science doesn’t have all of these answers, but it has far more than the DSM seems to realize. Much of that knowledge would have to come from fields like anthropology, evolutionary biology, animal behavior, and neuroscience. Indeed, one way to understand the project of Living Fossils is as an attempt to contribute to these operating instructions.

Without an operational understanding of the system whose errors it records, the DSM cannot reliably tell us what kind of failure we are seeing, where it comes from, or what kind of repair it requires.

Function from Malfunction

User manuals are useful documents because they indicate both how a thing should be used and how it should definitely not be used. This often makes them unintentionally hilarious. “Do not place child inside appliance.” “Do not use hair dryer while sleeping.” Civilization really is a struggle against human stupidity.

Unfortunately, the DSM isn’t nearly as funny—nor, in its current form, nearly as useful. But with a slight adjustment in perspective, it could be.

So far, we have been moving in one direction: from function to mechanism to failure. A washing machine exists to clean clothes. It does so with water, soap, and agitation. Those mechanisms create characteristic ways for it to fail. But inference can also move in the other direction: from failure back to mechanism, and from mechanism back to function.

Imagine you had to reverse-engineer a washing machine without being allowed to open one up. All you had was the complaint log from the customer service department: reports of leaks, mold, rust, torn clothing, short circuits, doors that won’t lock, and so on. You could infer a lot from that list. The machine probably uses water. It probably uses electricity. It probably moves clothing around. It probably has a door, a drum, a pump, seals, and hoses. A thing’s characteristic failures are clues to its design.

In a similar way, the DSM could have enormous value—neither as a medical map of natural diseases, nor as a clean taxonomy of mental illness, but as a collection of recurring symptom clusters. Properly understood, the DSM is not a parts diagram. It’s a complaint log.

That is not to say it can be entirely trusted. Many entries are based on self-report, which is not always reliable. After all, patients do not usually diagnose themselves with high blood pressure. Nor is the DSM a neutral transcript of human suffering. “Experts” decide what gets in, what stays out, what gets renamed, split, merged, or quietly buried in a footnote. That process involves science, yes, but also politics, fashion, bureaucracy, professional incentives, turf wars, and ordinary human error.

Still, the DSM is based on a great deal of observation across an enormous number of people. It is a fairly reliable record of the characteristic ways the human mind fails. And those patterns are clues to how it was designed—clues we might never discover if everything always worked properly.

Take auditory hallucinations. They may reveal something about how the mind tags its own contents.3 A thought is not just a thought. It is also marked, implicitly, as mine, yours, remembered, imagined, spoken aloud, believed, doubted, feared, wished, dreamed, and so on. If those labels fail, something generated inside the mind may be experienced as coming from outside it. The voice in one’s head becomes—just a voice.

The ordinary experience of reality, in other words, may depend on hidden tagging systems working properly—a possibility we might never have considered without the strange and terrifying error of hearing voices.

The path toward a better DSM begins with a retreat into modesty. Before the DSM can explain the mind, it should settle for accurately recording the mind’s recurring failures. That’s not nothing. A reliable record of breakdown is often the first step toward understanding design.

For example, rather than organizing disorders mainly around recognizable symptom clusters and clinical traditions, a better DSM might eventually organize them around the mechanisms involved. PTSD, phobias, social anxiety, and panic disorder, for example, might be grouped together not because they present identically, but because each involves some disturbance of the threat-detection system.

A better DSM would also include operating conditions. In addition to asking, “What symptoms does this person have?”, it would also ask, “Under what conditions do these symptoms appear?” The situation is not background information; it’s part of the diagnosis. Intense sexual jealousy, for instance, means one thing if someone’s partner is faithful and reassuring, and another if the partner is plausibly cheating or openly interested in someone else. The same feeling may be evidence of function or dysfunction. And if it is dysfunction, the situation helps determine what kind: malfunction, deprivation, or misuse.

Misery’s Silver Lining

Before ending, let’s return to the thought experiment about reverse-engineering a washing machine. If you had your druthers, wouldn’t you want as many customer complaints as possible? Wouldn’t you want the machine to have been used in all sorts of improper ways? You might not know it depended on electricity until someone tried to run it during a power outage. You might not know water could escape through the door until someone tried to run it upside down.

The modern world is doing something similar to the human mind. It is running ancient mechanisms under strange conditions and producing a thickening complaint log of distress. To me, this is one of the few silver linings of evolutionary mismatch: it can illuminate the design of our minds in ways a better-fitting world might not. Our failures may be among the clearest evidence of what we were built to do. People of the future, if they learn from us, may thank us for the misery required to obtain that evidence.

But one distinction must be kept firmly in mind. Washing machines and human brains were both designed, but they were not designed in the same way. My washing machine was designed intentionally, by humans at Samsung, over the course of a few years. My mind was designed blindly, by natural selection, across deep evolutionary time. Among other things, this means we have far more power to change washing machines than to change brains, at least for now.

Which means that if the people of the future want to act on the lessons our suffering provides, they will have an easier time changing the environment than redesigning the mind that has to operate within it.

The idea is a bit juvenile, but there’s a truth in it. We humans have certain appetites. When the environment allows us the opportunity to satisfy one of those appetites, we try to take advantage of it. So the environments we create for one another—such as the tools available to players of Mythic Quest—really matter.

My previous posts explored this idea when it comes to some peculiar ways we allocate stuff, including all-inclusive resorts, theme parks, and parking. If the environment allows people to reserve a lounge chair by the pool, they’ll do it. People like sitting by the pool. If the environment (of a video game) allows players to make a penis, they’ll do it. (I leave unanswered the mystifying question about why this creative impulse exists. I have no doubt finer minds than mine will have insight into this human peculiarity.)

This post explores cases in which some new ingredient was added to a society, looking at how human nature, often sitting latent, expresses itself in response.

Of Monsters and Men

In the 1980 film, The Gods Must Be Crazy, a member of the San tribe in the Kalahari finds a Coke bottle, discarded from an airplane. It turns out that this new object is useful for many of the San’s tasks, creating demand for the object. There being more demand than supply—just the one bottle—not to mention no price system to allocate it, conflict ensues. The solution is to throw the bottle off the edge of the Earth, launching the quest that animates the rest of the film. In this case, a new tool plugs into humans’ appetite to make life a little easier. The strategies the characters used to get access to the new tool were latent until the bottle arrived.

To take a second example from fiction before moving to the real world, the creation of The One Ring and the other rings of power—e.g., Nine for Moral Men, doomed to die—represents a similar case. Sauron took advantage of the appetites of the dwellers of Middle Earth. Of course men were going to slip the magical rings of power on their fingers, allowing Sauron to spring his trap, since he himself kept the “one ring to rule them all.” [SPOILER: It didn’t work, all thanks to the heroism of a rag-tag band of hobbits and their friends.] Of the three species—elves, dwarves, and men—we humans fared the worst, each ring-wearer becoming servants of the Dark Lord.

Ok, now back to reality.

Let’s start with fish.

If you were to watch piranhas, much of the time they are simply swimming along peacefully. But add some food, and their behavior changes dramatically. The ferocity is always there, waiting for the right conditions in the environment to bring it out. In this case, the right conditions include something they can eat nearby.

For humans, a crucial part of the environment is the set of rules that govern behavior.

In the first half of the 13^th century, religious conflicts were flaring across Europe and the relationship between the power of the state and the power of religious organizations was being negotiated. In Spain, the government used the Inquisition, authorized by the Pope, to assert its power.

In this context, new practices arose. For example, up to that time, if I accused you of heresy and I did not make my case, I faced potential punishment. With the advent of the Inquisition, officials were allowed to make accusations based on hearsay or circumstantial evidence, after which the accused would be encouraged to confess, using methods as well-known as they were barbaric.

The price of accusations was, under the Inquisition, sharply reduced, and the means used to “inquire” were such that once someone was accused, it was reasonably likely that they would confess or be killed. These new circumstances made the potential benefit of accusations much greater. This favorable cost/benefit ratio had the predictable effect. In Spain, estimates are that as many as 150,000 souls were prosecuted by the Inquisition. Evidence from the period suggests that accusations were an especially nifty tool favored by those of lower classes against the wealthy. Cheap, effective accusations could be used to settle old scores, compete with rivals, and so forth. Often, when property was confiscated from someone found guilty, the accuser sometimes stood to share in the bounty. The pattern repeated in witch hunts in Europe and, famously, Salem, Massachusetts. Again, leaders created an opening for attacks on rivals, leaving the accused defenseless because the charge was holding invisible beliefs or performing activities that left no physical evidence. (I like the Monty Python bit in which a character accuses a woman of turning him into a newt. Clearly being a human, and not a newt, he explains, I got better!)

In the case of the Inquisition, normal everyday psychology was sitting around, latent, until the environment changed. As Tolkien explored, humans are greedy and they crave power over others, to say nothing of our nearly insatiable appetite to accuse others of wrongdoing and revel in their ordeals of punishment. When ideas spread, they become like Coke bottles, a new part of people’s environments, potentially bringing to the fore previously latent appetites.

In some cases, these appetites are literal. Consider what happened as people developed animal husbandry and then factory farming. Throughout human history, animals have had no powerful lobby working to guard their rights, and the cost of eating animals dropped precipitously. The result is the 75 billion chickens slaughtered every year.

The results of other cultural innovations are more difficult to predict. If you haven’t watched it, I recommend The Social Dilemma, a documentary available on Netflix. It focuses especially on the innocent-looking-at-first “like” button. What happens when humans, with their appetite for attention, have a tiny new incentive to broadcast ideas others will like? The answer is the social media landscape we are all living in.

The introduction of new ideas does not always have such disastrous consequences. As the Enlightenment unspooled, an idea spread: people might have these things called “property rights,” which states might enforce. This idea got written down in places like the 4^th and 5^th Amendments of the United States Constitution and lots of people started believing it.1 The adoption of this idea led people who were living in places where the fruits of their labor could be protected to create things other people wanted. This practice had the happy consequence that more and more people can clothe, feed, and house themselves.

Along similar lines, about 2,500 years ago some people in Athens and then Rome thought that it might be fun if leaders were chosen rather than born. This idea, in various (and often imperfect) incarnations, leads people with power to do the sorts of things that are good for the people they rule rather than just themselves. Yes, that’s a naïve way to put it, but in theory democratic republics bring to the fore humans’ appetites to be liked, even loved, rather than feared, Machiavelli’s musings about this notwithstanding.

When I was a post-doc at UCLA, I was housed in the Anthropology Department, lucky to be around talented scholars such as Rob Boyd, Joan Silk, Susan Perry, Joe Manson, and others. I met with a certain amount of resistance from the University’s Institutional Review Board, which I discussed with my colleagues. It turned out that they had also had some difficulties—as have many others at various institutions—and from them I learned about an objection to their research methods I had never previously encountered.

Some anthropologists would go into the field, often an area where hunter-gatherers lived, to conduct experiments. People in these groups had very few material possessions. If an economist were to compute their per capita income, it would be less than one dollar a day. Hunter-gatherer participants would be compensated for their time taking part in the study, sometimes with tools such as fishhooks, which were in demand, and sometimes in currency that could be used in town.

The Institutional Review Board rejected proposals for some of this work in cases in which they judged that the compensation for participating in the research was too big. The argument was that the temptation to earn that much money was simply too great for their potential subjects to resist. Therefore, because the transaction was irresistible, it was coercive.

Up to that point—and, to be honest, still today—I was persuaded by Robert Nozick’s view about how to define and understand coercion, and I thought that coercion was a sort of conditional intended to undermine someone else’s will. In the classic case, I might impose a conditional on you of the form, your money or your life! What I mean is that IF you do not give me your money, THEN I will shoot you. Such conditionals are threats. They change your behavior, causing you to choose something you would rather not do and would not do if it weren’t for the threat.

Subsequent developments in philosophy extended and revised Nozick’s view, which some have seen as too narrow. There are arguments that certain kinds of offers, as opposed to threats, can be coercive. And Nozick’s view doesn’t exactly capture a dog on a leash: leading Fido here rather than there—keeping him out of striking range of the discarded chicken bone—seems coercive, but in a more straightforward physical way than the threat Nozick identifies.

Notice a big difference between the sorts of coercive threats the mugger makes and the allegedly coercive offer that the anthropologist is making. In the threat case, the target winds up doing something they otherwise wouldn’t want to do. In the offer case, with the intervention of the Institutional Review Board, the target winds up not doing something they would rather do. They are, from their perspective, worse off.

Why do administrators at universities want to prevent these “large” payments of six fishhooks rather than a single fishhook?

Think about the idea that offers can be coercive like the Coke bottle, a new idea to play with. In the case of the Institutional Review Board, with the framing that offers can be coercive, they can dictate the decisions made by people thousands of miles around the world, exercising their will over them from the comfort of their air-conditioned offices in Los Angeles. (If you think that this practice is acceptable, please let me know when your next job offer comes and I will decide what the maximum amount they can pay you without the offer being coercive.)

For these bureaucrats, this feeds several appetites all at once. First, Institutional Review Boards live in perpetual fear that their numbers will be culled if the people who propose research projects submit proposals that conform to all the rules. In such cases, there’s little for the staff to do, putting jobs at risk. Requiring revisions because it builds job security.2 Second, I can say from experience that the reports back from the Board often contain at least some aroma of moral superiority. You’re going to give them six fishhooks!? You monster! This sentiment is conveyed in standard bureaucratic language, but one can feel the finger-wagging.

And, of course, for at least some of them, there is the warm feeling of protecting those poor, vulnerable adults in Asia, Africa, or South America, who lack the bureaucrat’s sophistication and willpower to resist the siren call of a wage of six fishhooks per hour. The bureaucrat goes home feeling satisfied that she has chastised the powerful Principle Investigator and protected the hapless adult human being from whom she has removed the opportunity to choose whether or not to answer some survey questions for the wrong number of fishhooks.

Notice the theory of human nature that underlies this approach to policy. In prior posts, I’ve discussed the work of Ayaan Hirsi Ali, who gives sharp insight into the rules that govern certain practices in the religion in which she was raised, Islam. If you think of adult males as simply powerless to choose to behave when they see a whisp of a woman’s hair, then you might feel justified in draconian rules ensuring this temptation is never visible. According to this perception of people’s will, people making certain clothing choices are subject to stern punishment because they are being as coercive as compensating hunter-gatherers too much to complete a study. This view is a deeply paternalistic one, justifying why the few and powerful should choose for the many and powerless.

Don’t get me wrong. It is right and proper that Institutional Review Boards—and businesses and governments, and so on—try to prevent coercion in the sense of threats. A core ethical principle is that one person should not, and should not be permitted to, coerce another. And, of course, there are some people who aren’t able to choose, and are in need of paternalism. One prominent class of such people are called “children,” and we actually don’t want them choosing what’s for dinner, or every meal would be chicken nuggets and chocolate milk. We give people the vote at 18 in the United States, drawing a line about when we think their opinion about governance should count.

Humans like to make moral attacks. Expand the notion of “coercive” to include offers, and that’s what they’ll do. The “time to attack,” just like the time to penis, is vanishingly small.

Before concluding, I want to be clear that of course these matters are more complicated and it’s not quite as simple as saying that coercion by threat is always bad and coercion by offer is always fine.

Some threats are implicit. Nice bar you have here. Shame if something happened to it. In that case, there is no explicit threat, but the Nozick conditional—if you don’t give me protection money, then my gang will torch your bar—is in the mind of the speaker. It’s invisible, as all human thought is, but it’s also real.

The invisibility of these conditionals and the fact that they might be implicit can lead to much misunderstanding. Suppose a patron literally was just remarking that it was a nice bar and how sad it would be should it burn to the ground. When the owner serves the patron free drinks, we might ask: has the owner been coerced?

This ambiguity leads to a problem. Someone might believe a threat to be present when the speaker has no such intention in mind.

Shared understanding of what is coercive varies over time and from one community to another. This variation is inevitable because norms change over time and the topic is inherently fraught, to do with lots of stuff going on unseen in people’s heads. In one cultural moment, it might be just fine to say “it’s cold outside” as a way to persuade someone to stay inside, where it’s warm, a bit longer. In a different cultural moment, every offer by someone understood to have power might be able to be labeled coercive, and therefore subject to moral attack.

A very nice man saw that my partner and I were lost in downtown Istanbul. He offered to help us, and he provided impeccable directions to the Hagia Sophia, which turned out to be as beautiful as its reputation promised. The man invited us to have a look at his carpets for sale, just nearby. He had his young son serve us tea. He explained the care the women took in weaving the carpets and the dire conditions under which the women lived, on pennies a day, refugees from war-torn Syria. He explained that he would let these carpets, here, go for a steal of just 400 dollars, U.S., and which one did we like best?

Anyone who has been in such a situation knows how difficult it is to say no. These merchants are experienced and have enjoyed success for hundreds of years in their trade.

We walked out feeling a little bad, but with no merchandise and all of our money still in our wallets.

Did we feel coerced? I suppose we might have used that word if we had bought a carpet. Maybe. Certainly the merchant tried to persuade us. That seems like the right word.

But, in the event, we said no.

We had one powerful force on our side.

We’re adults.

If the ultimate goal of the 9/11 terrorists was to slow America down, to waste as much of our collective time as possible, they succeeded. Americans now spend, conservatively, hundreds of millions of hours per year in TSA lines, removing shoes, redistributing laptops, and explaining that the suspicious-looking item in the scanner is, in fact, a sandwich.1

In the midst of one such indignity, I had a thought: What if you could buy a “security-free” ticket? You could show up and proceed directly to your gate, arriving—just like in the old days—half an hour before departure. Of course, you’d be flying with other people who had also skipped security. Other modern safety measures, such as locked cabin doors, would remain in effect.

This is a hypothetical meant to convey a point, so don’t get lost in its feasibility. The important follow-up is: would you buy that ticket?

At first, my response was—Yes, by all means, now and forever yes. (These are the words Annie Dillard imagines the male praying mantis saying when asked whether he will mate at the cost of his head.)

But my second and final answer is—No, probably not. Especially not now that I have a daughter.

However, my becoming a dad is not the most significant factor here. 9/11 is. The events of that day showed that planes could be dangerous. Not that there was ever any real doubt beforehand, of course. We knew, in some abstract sense, that a plane in the wrong hands could become a weapon. What we lacked was a vivid example, something concrete enough to force a change. We lacked, as a psychologist would say, availability. So we did nothing—until something happened.

But here’s the thing: the risk on 9/10 was the same as the risk on 9/12—setting aside the impact of 9/11 itself2—yet we tolerated it without a second thought. Had anyone given it a second thought, we likely would have prescribed them therapy and medication for their “irrational” fear of flying.

Consider trains in the present-day United States. We “know” they could be used as weapons, but because, by and large, they haven’t been, we haven’t done much to prevent it. If something does happen, security will beef up overnight. (I can only assume nothing’s happened yet because even terrorists would prefer to avoid traveling on America’s trains.) So, for now, we buy security-free train tickets without a second thought, just as we bought plane tickets before 9/11.

The point is that we are often unaware of the risks we tolerate until some event changes our perception and renders what was once acceptable intolerable. This holds not only when the underlying risk remains stable, but even as it declines. After all, planes have become much safer since 9/11, independent of TSA security measures, yet my guess is that far fewer people would purchase a security-free ticket now than twenty years ago. Likewise, child abductions are far less frequent than they used to be, yet parents are more reluctant to let their kids roam.

Let’s get graphic for a moment. All else equal, as the world becomes safer, we should expect people to require less protection. Travelers should become more willing to forgo TSA security as planes become safer; parents should become more willing to allow independence as child abductions decline.

The graph should look like this:

Instead, it looks like this:

The difference between what we should see and what we do see arises because objective risk and perceived risk can move in opposite directions. The world may become safer even as it feels more dangerous.

What might cause this divergence? Availability, or how easily a danger comes to mind. And what tends to make dangers of all kinds readily available? The modern global news infrastructure.

News, as we know, tends to focus on tragedy—it has a “disaster bias.” As more of that tragedy is brought to our attention, two things happen:

1. Objectively, the world gets safer as people work to reduce the likelihood of similar tragedies happening again

2. Subjectively, the world feels more dangerous because people now have vivid examples of danger lodged in their minds

You’ll notice that technology is responsible for both sides of the coin. It makes pools safer while also bringing us stories of pool disasters. So how do we solve the problem of perception and reality being out of sync? You guessed it: more technology!

The Let Grow website, for example, tries to restore the parental confidence that several decades of true-crime television helped erode. It gives parents what they now need—research, pamphlets, even legal backing—to feel comfortable doing what a parent two generations ago would have done without a second thought: let their child go outside and play.

There’s even a Kid License!

Availability isn’t the only factor shaping our tolerance for risk. Returning to my own answer about whether I’d purchase a security-free ticket, another factor is changing stakes. Now that I have a daughter, I have more to lose and less to gain by taking risks. I am also not as young as I once was, so it may take me longer to recover from a loss. Finally, as Jean Twenge notes in Generations, it’s rational to tolerate less risk as the world becomes safer. It was one thing for pioneers to tolerate rattlesnakes when greater dangers were afoot (sorry), but it’s now foolish for someone to wander among rattlesnakes when they don’t have to.

So there may be two forces at work: some risks feel larger because they are more available, while others become less worth taking because we no longer have to take them. The open question is whether, in reducing risk, we have also reduced reward.

I’m so tickled with polls that I’d like to end with one more. When I first drafted this piece, I assumed a security-free ticket would be less expensive because security is a value-add. But then my wife made a good point: many people experience security as pure inconvenience and would happily pay to skip it. What do you think?

Party Polarization—RK

Here is the way that human political psychology doesn’t work.

I go about my day and learn this and that. I listen to my parents, I read books, I take classes, that sort of thing. As I get older, I choose an Ideology with Principles, with capital letters and all that. So, you know, my Ideology is Conservative and I believe that people should pull themselves up by their own bootstraps and that the best thing the government can do is get out of the way. Or maybe I’m a Liberal and I think that the government is the solution to helping the least among us and that’s what really matters.

Or whatever.

Now, whenever someone asks me my position on a specific political issue, I consult my Ideology and take the position consistent with it. If I’m a Conservative and someone asks me if I believe that that government should distribute wealth from the rich to the poor, I say hell nah.

In this world, which doesn’t exist, we would see predictable patterns in data. Take, for instance, the General Social Survey, a poll that has been conducted with similar questions since the early 70s. The key pattern we would see is that people’s positions on a variety of different questions would be correlated. So, for example, people who are Ideologically Liberal would give the same answers to questions about, say, affirmative action and income redistribution—yes please to both—and people who are Ideologically Conservative would give the same answers—no thank you to both.

If ideology drove views on particular issues, we would see nice big correlations across different issues.

My former collaborator Jason Weeden over at his Substack, Charty Arps, explores such datasets and how things have changed over time.

Consider this graph from this post:

In 1996, the correlation between people’s views on abortion and income redistribution wasn’t just small, it was zero. Prior to that, it was a little bit negative. In 1996, if you asked someone their view on abortion, you would get no insight at all about their position on income redistribution. These two policy views were completely unrelated.

Here is the way that human political psychology actually works.3

When I have to take a position on some particular policy, whether because it’s put to a vote, a pollster asks me about it, the topic comes up in a conversation, or whatever, I ask myself two questions.

1. How does this policy affect my interests, understanding my interests to include my income, my wealth, my reproductive success, my freedom, and so forth?

2. What will people around me think if I adopt one position or the other?

Sometimes these two questions will point in the same direction. I’m a rich White person in Alabama in 1985 and someone asks me about my position on government wealth transfers. These transfers (1) reduce my wealth, and (2) people around me don’t like the government very much so I’m agin’ it!

Now, if I’m a poor White person in Kansas, say, then maybe that policy would help me in terms of my wealth, but I know that my neighbors oppose it. Then I have to balance the strength of the forces pulling me in opposite directions.

The power of these two considerations changes depending on the issue and the context. In some cases, I might be relatively unaffected by an issue—say I’m retired and affirmative action doesn’t really affect me one way or another—and in some cases other people might not care much about the topic in question—the trans issue in 1996, for instance.

Now, back to the graph. For some reason, starting around 2014, the correlation between people’s positions on abortion and income redistribution began to shoot up.

What might explain this?

One possibility is that ideology started driving people’s opinions. Remember that in the hypothetical world that I think doesn’t exist, people derive their specific positions from general principles.

That’s possible.

Here’s a different explanation. Around 2014, the second driver of people’s opinions—what others think about me for my positions—increased because social media made these opinions way more visible. In 1996, you couldn’t post a black square or a green circle to show which way you leaned on a topic because in 1996 people didn’t have Twitter accounts.

So, as the cost of having an opinion that differed from that of one’s community increased, the correlation between different opinions increased. More generally, as the second force—the importance of sharing the same view as others in the community—increases relative to the first force—the power of self-interest—we should expect to see more and more homogeneity in people’s political views.

Subscribe to Charty Arps for more.

An easy mistake is to judge them by today’s standards, a practice sometimes referred to as presentism. I find presentism naive because it forgets how much courage it takes to swim even a little against the current. For example, how many psychology undergraduates, when prompted by their textbook, actually face the wrong direction in an elevator? My guess: not many. Even violating this low-stakes social norm is too uncomfortable, let alone something with real moral weight.

In today’s moral climate, it has even become fashionable to condemn those who moved in the right direction for not moving far or fast enough. (You’ll have to imagine how hard I’m rolling my eyes.) Thomas Jefferson is a common target. Sometimes his slaveholding receives more attention than his role in articulating ideas later used to undermine slavery itself.

Another easy mistake is to veer in the opposite direction and maintain that morality is purely culture-bound and therefore beyond judgment—a position known as moral relativism. “It’s their culture,” the argument goes. “If it involves female circumcision, infant sacrifice, or human trafficking, what kind of moral superiorists are we to say it’s wrong?”

Fortunately, we needn’t take the moral relativists that seriously. In a sense, they do not even believe what they’re saying. Out of one corner of their mouth they claim we cannot judge cultural practices like female genital mutilation, yet would you ever catch one admitting that the South was just as moral before the Civil War as after it?1 There is something strange—by which I mean hypocritical—about the moral relativist position.

So how should we judge people of the past? One useful technique comes from sports. To answer how good a player from a previous era really was, analysts now rely on “era-adjusted statistics.”

Before era-adjusted statistics, people could get away with saying something like: “Sidney Crosby is great, but he’s no Wayne Gretzky. Hell, Gretzky scored 92 goals one year. The closest Sid’s ever come is 51!” (These are hockey players, by the way.) Of course, this barstool observation ignores the fact that during Gretzky’s prime—the 1980s—the average goals per team per game was 3.83, while during Crosby’s prime—the 2010s—it was more than a goal lower, at 2.75.

The claim that Gretzky was a cut above Crosby overlooks another fact: Crosby skates faster and shoots harder than Gretzky ever did. The All-Star Skills Competition’s “Hardest Shot” event, for example, shows the winning shot speed rising steadily by decade—from 98.3 mph in the 1990s, to 101.7 in the 2000s, to 105.8 in the 2010s.

If Gretzky was the better player, why didn’t he skate faster or shoot harder? Granted, skating faster and shooting harder don’t necessarily make someone a better hockey player—Gretzky was known primarily for his ability to anticipate the play—but the discrepancy in raw skills still weakens the claim of his superiority.

The answer, of course, is technology. Crosby has access to better equipment and better science than Gretzky did.

So how do we compare the two? Against the average of their era. A better metric than “Gretzky scored almost twice as many goals as Crosby in their best years” is that during the 1980s, Gretzky’s goal production accounted for roughly 20% of an average team’s scoring, while Crosby’s accounted for about 17%. So, as goal scorers, they weren’t too far apart.2

Can we apply the same logic to morality? I don’t see why not. The question isn’t whether Jefferson met today’s standards, but how far he rose above—or sank below—the moral average of his time.3 I would say he rose above it by helping articulate a vision of greater freedom than had previously been imagined. He advanced the moral baseline not by freeing slaves, as we might now wish, but by crafting language that later generations would use against the institution of slavery. Indeed, the idea that every person should be treated equally under the law is such a fixture of our moral understanding that we feel comfortable skewering Jefferson for not getting there faster—even though pushing the idea further in his own time might have consigned both him and his ideas to obscurity.

Meanwhile, what are the moral equivalents of better equipment, training methods, and sports science? Technologies that make moral transgressions costlier and moral signaling more rewarding. Social media, for instance, makes violations of prevailing morality more transparent and increases the status one can gain by seeming righteous. Such technologies do not necessarily make people more moral so much as they alter the payout structure, making virtue more lucrative to signal and vice costlier to display.

“Moral progress,” then, reflects not nobler souls, but different incentives made possible by new technology. Rediker hints at something similar when he notes that abolitionism gained decisive force only after the Brooks slave ship diagram made the horror easily visible. Before then, abolitionism lacked, as Rob would say, a coordination point. In short, we should no more assume we are morally superior to our predecessors than assume Crosby is a better athlete than Gretzky.

“Era-adjusted morality,” then, allows us to honor Jefferson’s achievements and contextualize his shortcomings. Not only does this approach help us judge past figures more fairly, it can also help us take more realistic moral stances today.

Which brings us to one final question: why judge the dead at all? They certainly don’t care. We must do it for ourselves—either to jockey with one another for status or, more loftily, to calibrate our society’s progress and ask what direction it should take next.

Rediker, for instance, uses his moral accounting of the slave trade to argue for reparations. The next step in moral progress, he suggests, is to repay those whose families were ripped from their homeland and forced into servitude.

To me, this is right in theory but wrong in practice.4 Such a policy would likely divide the country even further, setting us back in the same way that Jefferson’s premature attempt to abolish slavery might have stalled the nation’s founding altogether. This is, of course, a position I hope those of the future judge me fairly for, as someone who still hasn’t faced the wrong way in an elevator.

This post continues along the same lines.

Why am I so focused on prices?

Two reasons.

First, I believe prices often make people better off. And I’m in favor of that.

Second, I think prices contain a lesson about consciousness.

Let me start with a story.

When I was working on my Master’s of Public Administration (MPA) at the Fels Institute of Government, I was lucky to meet another student, named Adam.

Adam worked in the non-profit sector. He directed a facility for young men who were charged with a crime and, instead of being incarcerated, were diverted from the criminal justice system. The facility received money from the government—mostly the City of Philadelphia—and the facility provided the kids in the diversion program with the necessities of life and services designed to support them. These services included activities, group psychological counseling, and individual therapy.

The Master’s program was small and intense, so we all got to know each other pretty well. Adam frequently shared his frustration with the system surrounding juvenile diversion.

His frustration wasn’t that these programs lacked funding. There was a steady stream of money from City Hall.

But the kids didn’t have great outcomes. The key measurement for kids in diversion is whether they were arrested again. And, indeed, many of them became involved in criminal activity not long after leaving the program. Further, tragically, more than one of the kids Adam came to know would be killed as a result of violence. A worse outcome there could not be.

The weakness of the diversion program was mysterious. If you talked to the people involved, most of them had good intentions. The people in City government genuinely wanted to help kids and spent money to try. They were motivated if for no other reason than they wanted to be able to point to policy successes when they ran for re-election. Similarly, the people who ran the programs had good intentions. You had to have a certain public-spiritedness to work in the non-profit space, which was often frustrating and less remunerative than the for-profit sector. Why were all of these good intentions and funds not turning into good outcomes?

I got the chance to watch Adam analyze the system in which he was embedded as he worked through the coursework in the MPA program, especially the courses on economics and program design.

Here’s how he started to view the problem.

You have kids who receive services. These services might or might not be useful to them. But if the services aren’t useful to them, they have no way to convey this—by, say, choosing a competitor’s services. The City, or other entity, makes that decision for them. So the feedback loop from consumer to seller is broken. Organizations are selling to the government, but the government isn’t consuming the service.

As a consequence, the service providers are not incentivized to provide the best service they can for the money they receive because they have a, well, captive audience. Instead, they have two other key incentives. The first is compliance. To keep getting money from the City, they must follow many rules: employees have to have certain certifications, a certain number of hours must be spent per child, that sort of thing. The second key incentive is to keep costs down. They don’t earn more money by having high-quality services that make kids want to go there because the kids aren’t making the decisions. They can’t improve their top line by increasing consumer demand. But the providers do improve their bottom line by keeping costs down.

This set of incentives is, of course, very different from the set that businesses face. Let’s take the example of a new paint-your-own-pottery studio that everyone in the greater Philadelphia area should absolutely visit. At the studio, we care a great deal about the quality of the experience that we deliver (4.9 out of 5 stars on Google). Everyone who comes to our studio could, instead, consume leisure time elsewhere. We have made adjustments over time as we’ve learned about the market. Further, just like other sellers, we update prices of specific items depending on demand as revealed by customers. Unicorns are popular. We raised the price. Chickens are not. On sale now! By looking at sales data, we have a window into what customers want. Their purchasing behavior reveals their preferences.

Adam came to believe that the diversion system was failing kids in part because their preferences are invisible. No one knows what the kids prefer because they can’t choose one service versus another. The consequence is programs that, excuse my French, suck. Service providers aren’t punished by the market in the way that our lagging demand for chickens in the pottery store caused us to lower our chicken prices.

After the all-inclusive and Disney posts, I spoke to a reader about how much I liked prices and they said, “Do we really have to extract every dollar from consumers?”

And that’s fair. Pricing is generally designed to maximize profits.

But that fact doesn’t mean that only sellers are better off with prices.

As we saw in the prior posts on all-inclusive resorts and theme parks, when goods and services aren’t allocated with prices, you get weird results that make consumers worse off, too. Poolside chairs go to the people most willing to break the rules about reserving chairs. Slow service plagues those at the swim-up bar.

Back in the real world, away from luxury travel, one of my favorite examples, thanks to the late Donald Shoup (see his book The High Cost of Free Parking) is the more mundane case of street parking in the United States. Parking on the street is often free and, if not free, then well below the market-clearing price.1 Shoup pointed out that the result of low or zero prices is cruising. Drivers circle, looking for an open spot, sometimes creating 30% of the traffic in an area. Everyone pays an extra cost in time, fuel, and frustration, rather than in dollars. And the parking authority learns nothing. When San Francisco ran an experiment and let meter prices vary block by block to target one open space per block, while some prices climbed, prices on many blocks fell because demand had been overestimated. Cruising dropped, drivers got their spot reliably, and reported being better off. They had been paying for parking all along. They just weren’t paying in a currency that told anyone anything because their frustration and cruising was invisible to the people setting the prices.

Concert tickets illustrate a related but distinct pathology. Ticket prices are often listed far below what people would actually pay. Many people express warmth about this. Isn’t it great that artists are letting their less-well-off fans enjoy their music? Well… When Taylor Swift’s Eras Tour tickets went on sale at a face value of around $400, the market-clearing price for good seats was several thousand dollars. And that’s what the tickets ultimately sold for. But people didn’t buy them from Taylor Swift. Instead, these margins went to scalpers, and the resale platforms everyone hates, including StubHub and Ticketmaster’s “verified resale” operation. True fans either paid the premium through a middleman or watched bots clean out the inventory in the first thirty seconds and didn’t get in at all. Now, that’s not to say that Swift didn’t do very well, only that third parties captured a lot of the profits. (To her credit, Swift generously distributed the bounty to staff). The tickets ultimately sell for what people will pay; the only question is who collects it.

To return to Adam, his stories about his experience in the rarefied world of non-profits bounce between comedy and tragedy. He told me about the time he attended a juvenile criminal justice conference. The gathering ranged from law enforcement to politicians to judges to therapists and more. When his turn came to speak, he began with a question for the audience. Raise your hand, he said, if you would run your company, department, or program just the way that you currently do if you were paid based on how well it worked, keeping people out of prison a second time. Not only did no hand go up, but he was met with raucous laughter.

Something is rotten in Denmark.

Adam is developing workarounds because the system as a whole is very difficult to reform. Here’s one idea: how about asking people in the system what they need? People in the justice system generally can’t vote with their wallets, but that doesn’t mean they can’t get a vote. He has piloted a program in which—gasp!—he asks people what they need to stay out of trouble down the road. If you are going to be released in six months, what do you need to get a job, find a place to live, support yourself? What is your biggest pain point? How can the resources available be used to solve this problem?

Adam’s efforts are not exactly a vindication of pricing. But they point to the general idea that the absence of prices in many contexts presents problems, and not just for one side of the market or the other. The fundamental reason is that information isn’t available. When you see our unicorns cost $25 and our chickens cost $20, you infer something about the demand for unicorns.2 Prices in markets represent the way that sellers have responded to the information they get from consumers.

Adam’s surveys, like prices, also capture information. As do reviews on platforms such as Google and Yelp. People who produce goods and services meet demand best when they have information about that demand. Often that’s buying choices and prices, but not always.

When you think about prices, it’s easy to get stuck on the transaction, the amount of money. And it’s easy to think, hey, if now there’s a price that wasn’t there before, stuff got more expensive. But to focus on prices per se is to miss the bigger picture. The bigger picture is that the situation overall often gets better, for everyone. The problem with services to youth offenders in this case might not be how much—or how little3—we spend on it. The real problem is information. All of the frustration these kids feel with their services is lost to the system.

As Adam did, we can try to solve this problem is to make the information visible through surveys, but surveys present their own problems. One is that people lie. Another is that surveys cost money to implement. A third problem is that some people in the system might have a stake in hiding the information they might reveal—think about low-quality service providers. Yet a fourth problem is that good survey design is hard.

Not to take anything away from Adam’s efforts. It’s just that the price system is a really clean way to elicit people’s true preferences. There are others, but all of them, including voting, have their drawbacks.

What are some other areas in which price information is absent? In the prior post, I talked about healthcare. I’m not saying prices will solve the many problems of that complicated space, but consider all the information that is lost because people don’t pay healthcare providers directly. If you are skeptical that medical care can be a normal everyday market with prices, have a chat with a veterinarian.

I live in Philadelphia, a city known for “eds and meds,” the two sectors that make the local economy spin. There is no shortage of institutions of higher learning or medical providers. I recently wanted to book an appointment with a specialist and contacted the big players in the area. The first available spot was three months away. As I was writing this post, a friend of mine told me about her need to see a medical specialist. She wasn’t able to book an appointment at all, but is now on waiting lists to get one.

I am reluctant to get into another market in which consumers don’t pay prices, but public education is another. The research in this area is complicated—the linked paper finds that a 10% increase in spending yields a 0.07 standard deviation improvement on educational metrics. This finding reminds me of what Adam described. There’s lots of money washing around the system, but there’s no systematic benefit—or not much systematic benefit—to throwing more money at the problem. Could this be because there is a loss of information, with students having no way to vote with their feet and parents having limited ways to vote with their wallets?

Related, as someone taking care of taxes and compliance for a small business, I can tell you one additional area in which there is deep, profound market failure. The various web portals that businesses must use—I have in mind Pennsylvania’s MyPATH system and especially the state’s unemployment insurance management system—are, in a word, an abomination, an offense against god and nature. I have lost years of my life simply trying to log in to these sites,4 never mind trying to satisfy all of the requirements for filing and paying all the fees and taxes. My displeasure cannot be voiced by opting out of using these systems—the state will take away my property and freedom if I do—and I cannot use another provider, as the state claims a monopoly. I have bountiful information I would love to convey to the designers of these platforms, and I hope to meet them in a dark alley someday to share my thoughts.

/rant

What does all of this have to do with consciousness?

Aristotle, in On the Heavens, wrote that “a man, being just as hungry as thirsty, and placed in between food and drink, must necessarily remain where he is and starve to death.” What is a person—or a donkey—to do?

Let’s suppose that Marvin Minsky more or less got it right when he suggested there is a Society of Mind. Imagine that there are all these different little systems in your brain—whether you call them “modules” or something else is up to you—and they go about their business all day, tending to their little domains of responsibility. One little system is designed to make you want to eat, one wants you to run away from that bear, one wants you to sit back and relax… but, ultimately, you can’t pursue all of these goals at any given time.

How is that decision to be made?

Somehow, there has to be some way to compare how urgent it is to eat and how urgent it is to flee the bear. If the predator-evasion program and the get-food program both say, yeah, my thing is important right now, what’s an organism to do?

This problem is why these different systems all need to have some sort of common currency. It’s 57 units urgent to flee, but only 8 units urgent to eat. My guess is that, somehow, all of our motivational systems need to express their urgency with some kind of currency that can be compared. In our experiences, this is felt as the intensity of a feeling or an emotion.

The dollar in your wallet can only be used to contribute to the payment of one thing. You can use it to buy a Big Mac, a ticket to Reno, a new laptop… The reason that modern economies work as well as they do is that we have found a common metric—currency—to price these varied goods and services. You can only do one thing at a time and your dollar can only be put to one use.

Prices tell us which things are more valuable than others and the intensity of feelings tells us which priority is more urgent than others. Consciousness is the currency of decision making.

In Walden, Thoreau writes: “The greater part of what my neighbors call good I believe in my soul to be bad, and if I repent of anything, it is very likely to be my good behavior.” By now, it won’t surprise you that I feel something similar about my own field: most of what my colleagues call true I believe in my head to be false, and if I repent of anything, it will be apparently be keeping my mouth shut about it.

I sometimes wonder which is stronger, though: my instinct for critical thinking or my appetite for rebellion. Nearly everything I’ve thrown myself into, I’ve eventually torn down. My disillusionment with psychotherapy today mirrors my disillusionment with the academic study of literature ten years ago and may well mirror my disillusionment, ten years from now, with writing itself. (God help me then.)

Of course, underneath this criticism and disappointment is an abiding passion, otherwise I would just leave these things alone. My feelings echo those of Paul Minot, who wrote of his field of psychiatry: “I’ve been practicing psychiatry for 38 years. I love my job, my peers, and my patients. But I’ve come to the conclusion that I’m participating in the biggest intellectual scam of this era.” Similarly, despite believing less and less in the basic tenets of my field, I still love my work and my clients, and still cling to the promise at the center of therapeutic work: living the good life.

Anyway, the reason I’m navel-gazing at this particular moment is that I recently read two books by Stephen Grosz—The Examined Life and Love’s Labour—which cemented two truths that had been forming in my head. The first is that the portrait of human life suggested by psychoanalysis and other psychotherapies is often beautiful and compelling. Grosz’s pages are filled with stories from more than forty years of practice, stories capable of softening even the most hardened reader. I hope never to become so dead that I am unmoved by such writing.

The second truth is that much of it is wrong. Psychotherapeutic narratives, in general, don’t seem particularly concerned with being true—despite a good deal of chest-beating to the contrary—and are often closer to stories than to scientific explanations. Grosz’s interpretations of his clients’ predicaments are filtered through a distinctly psychoanalytic lens, one that frequently renders reality more profound and complex than it is.

Here are a few examples.

In Love’s Labour, Grosz interprets a client’s preference for older men as an unconscious attempt to find “a man who will be the mother I never had.”1 A more straightforward explanation is the evolutionary one: women tend to be attracted to status, which older men typically have more of, because over our long history, status tracked a man’s ability to provide resources and protection.2

In another section, Grosz writes: “To make sense of how and why we have sex the way we do, we must look closely at our personal history, especially our earliest relationships… In a way, our sex lives can be thought of as a solution to the problems given to us by our earliest fears, longings and animosities.”3 Here again, personal history is privileged over current situation—a default focus I have challenged at great length.

More to the point, wouldn’t it be simpler to say that to make sense of how and why we have sex the way we do, we must look closely at our species’ history? Our sex lives are, at base, solutions to the problem of genetic transmission—to the evolutionary imperative of passing our genes into the next generation—rather than to early, unresolved feelings. Early relationships and experiences can shape the details, of course, but they don’t set the terms. I see no reason for prioritizing them over the most fundamental driver of sexual behavior across the animal kingdom.

Grosz describes another client, Emma, who began seeing him “because she’d become acutely depressed after beginning a Ph.D.”4 More interesting than the interpretation Grosz eventually offers is a different question. Is there currently any room, in the field of psychotherapy, for explanations like this:

As Shani says, “Being a graduate student is probably one of the worst ratios of stressors to coping skills. When you’re a grad student, you’re poor, ignorant, stressed, and still immature.”

In discussing another client, Grosz writes of the common tendency to resist change: “We hesitate, in the face of change, because change is loss.”5 Not really. It would be more accurate to say that change is risk. Conservatism in behavior and morphology is widespread across the animal kingdom for the simple reason that there are far more ways for things to go wrong than right. People are loath to abandon something that is working well enough because, on some level, they recognize that it’s already a hell of an exception in a cold, unforgiving universe trending toward entropy.

Let’s spend another moment on this point. Therapists are often confused about why clients don’t change, overlooking that the client’s situation—which seems, and is often presented as, unsustainable—is actually working well enough. A therapist might wonder: “Why does she insist on working herself so thin that she keeps getting sick?” The client might even ask the same. But much of the time, the answer is that this is the client’s best available option, given the full set of constraints and tradeoffs.

Another of Grosz’s clients is boring. Grosz writes of boredom:

Boredom can be…a sign that the patient is avoiding a particular subject; that he or she is unable to talk directly about something intimate or embarrassing. Or it can mean that patient and psychoanalyst are stuck; the patient is returning again and again to some desire or grievance that the psychoanalyst is failing to tackle. A boring person might be feeling envious, and might kill a conversation…because he cannot bear to hear a helpful or compelling idea coming from someone else. Or the boring patient may be playing possum—just as there are beasts in the jungle that survive by playing dead, some people, when frightened, simply shut down. It’s also true that psychoanalyst and patient will sometimes unconsciously collude to desiccate the atmosphere between them because they fear things becoming too emotionally disturbed, or too exciting.6

Counterpoint: Some people are boring. Others are short, near-sighted, uncoordinated, or extremely sensitive to noise and light (nice to meet you). Boredom more likely reflects ordinary variation than hidden meaning.

As a final example, one of Grosz’s patients, Elizabeth, continued going out and living her life while her husband was at home dying. Elizabeth also

…found it more and more difficult to touch her husband, let alone have sex with him. Frightened of his death, reminded of her own mortality, and angry about being left to someday face her death on her own, Elizabeth found herself rejecting her husband in his final months.7

Or—Elizabeth didn’t want to have sex with a dying man. Sadness and grief are often enough to extinguish desire, let alone the signs of death and decay.8

Now, Grosz’s interpretations in these cases could certainly be correct. I don’t know these clients; I haven’t worked with them. Then again, Grosz is attempting to describe human behavior—a subject studied by many disciplines. While those disciplines don’t stop applying in the therapy office, they often stop being applied. In Grosz’s pages, there is a clear preference for the assumptions and narratives of psychotherapy, and a corresponding neglect of alternative explanations.

By “alternative explanations,” I mean not only the evolutionary and situationist perspectives I have advocated, but also simpler, common-sense, and statistical ones. Isn’t it much simpler to say that a boring person is boring because people vary on every observable trait than to presume that the client is playing possum? Doesn’t common sense suggest why someone might not want to have sex with a dying partner? And doesn’t a statistical approach—Bayesian reasoning—lead us to presume that, all else equal, the graduate student is depressed because she is a graduate student? Why are these interpretations so often overlooked?

“All else equal” is the key phrase. Grosz’s interpretations are not so much wrong as unlikely and indirect. My reaction to many of them is: “Could be…but probably isn’t. And even if it is, there’s a more direct explanation.”

To be fair, psychotherapy often deals with outlier cases—otherwise clients would not seek such an involved process—and no one writes about typical clients. So, the simplest explanations are less likely to apply to the cases Grosz selects. But this doesn’t tell the whole story. The bigger issue, I think, is that Grosz is wedded to a particular version of reality—in his case, a psychoanalytic one—which inevitably shapes what he sees.

Of course, this is a human tendency, not a Grosz one. In order to make adaptive decisions, we must attend selectively and interpret quickly. Everyone has some bias, then—some built-in way of seeing. But it’s possible that my field has a particularly skewed version of reality. For whatever reason, psychotherapy is full of interpretations that are compelling, profound, and largely wrong.9

If I am right that Grosz is wrong about many of the causes of mental distress, you might expect this to undermine his effectiveness as a therapist. I am sure that it hasn’t. Indeed, I’d be willing to bet 100 subscribers that Grosz has helped his patients tremendously. Nor do we need to guess: a large body of research suggests that psychotherapies of all stripes produce broadly similar outcomes, despite having widely varying foundational assumptions and techniques.

In my view, this is because the content of a particular therapy matters far less than the process of someone listening and caring. Grosz’s books themselves are a testament to how well he listens and how much he cares. At the most fundamental level, an interpretation is a signal to the client that the therapist is trying to understand, and that alone is often enough.

But even at the level of content, stories are typically more helpful than truth. Clients don’t want an accurate description of their misery as much as one they can use. And the same goes for therapists. The client gravitates to what will heal them, the therapist to what will make them the healer. Therapy becomes an attempt to arrive at an explanation that satisfies both. A diagnosis, for example, can legitimize a client’s otherwise confusing or unacceptable behavior and, at the same time, bring their healing more firmly under the guidance of the therapist’s expertise.

Just like that, both can heal.

Seen in this light, the interpretations Grosz arrives at begin to make more sense. He and Emma can do more with an explanation that locates her depression in early childhood dynamics than with one that points to the practical struggles of Ph.D. programs. He and his boring client can work with the idea that the client is playing possum—they can ask why, for example, rather than shrugging their shoulders and muttering c’est la vie. An actionable interpretation, whether true or not, solves one of the most difficult parts of suffering for both client and therapist: not knowing what to do.

The same dynamic takes place at a larger level. Not only the therapist and client, but clinical psychology and culture, collude to arrive at stories that serve them both. While these stories are not science, presenting them as such makes them more effective. Psychotherapy is less a science that occasionally uses stories, then, than a storytelling system that occasionally borrows the language of science. Movements like “evidence-based therapy” are simply part of the performance, part of the ritual of healing, which benefits from the appearance of rigor and expertise. That is why, to an extent, it doesn’t matter what the evidence is—or whether it exists at all.

So, therapeutic explanations work not because they are true, but because they are useful. Yet we often treat them as true to preserve their usefulness. If we expose therapy as a kind of useful fiction, we risk breaking it. But if we don’t, we continue pretending it’s something it’s not. Perhaps we can’t have both: a field that heals and is honest about why.

I think this explains the sometimes jarring duality of my writing and practicing lives. In session, I have little difficulty deciding whether to communicate my stance that, for example, ADHD isn’t real. It is usually clear whether the client will benefit from that perspective or not. But when I write about ADHD to a broader audience, it’s a different story. There, I want to be as truthful as possible; I have a responsibility to do so. Yet readers may be using ADHD in a variety of ways, some of which will be punctured by my little javelin of truth.

Which brings me back to navel-gazing. Despite how it might seem, I am partly ashamed to be writing this. I am not wholly comfortable exposing my field. It makes me feel like the sort of person who turns off the music at a party and asks whether anyone has actually been listening to the lyrics. For many, if the beat is good—if the story works—people will dance all night.

I’d like to dance, too. Hell, I got into this field because, while unemployed, I became enraptured by the writings of Freud and Jung. I was reluctant to admit that early on, partly because Freud was passé and partly because everyone else seemed to have chosen the work for more practical reasons. But the truth is, I wanted to see the world the way these original masters did; I wanted my working days to be like theirs.

Not only did I want them and their descendants, like Grosz, to be right—I wanted to be one of those descendants. I wanted to carry the flag forward. Their versions of reality were richer, deeper, more meaningful than the one I now find myself describing.

It’s only that they were, by and large, wrong.

• A key relationship pattern is brought into awareness and resolved (transference)

• Unconscious dynamics are made conscious

• The therapist provides the client with tools and techniques to better manage their thoughts and emotions

• Insight, or self-awareness, leads to maturational change

A simpler explanation is that the client (a social animal) forms a meaningful social relationship with their therapist, one that conforms to an unparalleled degree to the client’s wishes.

Ignoring emotions is harmful because:

• They won’t go away until they’re brought into the open and fully processed. Until then, they may cause you to act out in unproductive ways.

In reality, ignoring emotions can sometimes help. They often pass on their own and aren’t always worth trusting—especially in the modern world.

People are self-critical because:

• They’re insecure

• They’ve internalized critical voices from early relationships

• They judge others and assume they’re being judged in return

A simpler explanation is that it’s better to beat ourselves up than have the world do it for us.

You meet the criteria for the following diagnosis: [insert].

A simpler explanation is that you vary across psychological traits—gullibility, impulse control, empathy, conscientiousness—and live in a world vastly different from the one in which humans evolved.

Now that we’ve discussed your diagnosis, you’re starting to notice how true it is. That’s a sign we’ve gotten it right.

A simpler explanation is that suggestion and confirmation bias are a powerful combination.

Love’s deepest motivation is:

• To find the opposite of the love you received as a child

• To recreate the love you received as a child, for better or worse

• To resolve unfinished business from early relationships

A simpler explanation is that love motivates us to pair up with someone of equal or ideally greater mate value. Mate value is based on different things for men and women—typically status for men, beauty for women.

People struggle to find love because:

• They don’t love themselves

• Their early relationships were inadequate

• They are afraid of losing themselves

A simpler explanation is that even for those of moderate to high mate value, finding a long-term partner is difficult because of the duration and intensity of coordination required.

People who pursue sex actually want:

• Connection

• Power

• Validation

• Excitement

A simpler explanation is that an animal whose ultimate evolutionary purpose is to procreate occasionally takes measures to do so.

Men who sleep around do so because:

• They’re insecure or compensating for something

• They are afraid of genuine connection or commitment

• They were taught a form of toxic masculinity growing up

A simpler explanation is that males across many species pursue multiple partners when they can.

These two events happened together for a reason.

A simpler explanation is that not everything that happens together means something. It might be coincidence.

Dreams are meaningful.

A simpler explanation is that dreaming likely has a function, but we don’t know what it is. The content may just be noise.

In 1993, Sally Ride, the first American woman in space, referred to her trip on the Space Shuttle Challenger as an “E-Ticket ride!”

The background for this reference dates back to 1955, when Walt Disney opened Disneyland in California. At that time, the current pay-one-price for all-you-can-ride model would be more than 25 years away. Instead, guests could buy tickets or books of tickets—labeled from A to E—for different attractions. The more mundane attractions required an early-alphabet ticket. An A-Ticket got you on the Horseless Carriage (oooh!) or the King Arthur Carousel. Rocket to the Moon or the Jungle Cruise required a precious E-Ticket.2

When the normal price system isn’t being used, another method for allocating scarce resourced must be. There are only so many guests who are going to ride Dumbo in any given hour—about 400—and only a few of them are going to get the exit row seat on the aisle on the right side of the plane, my personal favorite. Just like everything else in life, spots on attractions are a scare resource that must be allocated somehow.

In the 80s, Disney in the U.S. moved to an all-you-can-ride system for the attractions. (The ticket system persisted at Tokyo Disneyland all the way to 2001.) Spaces on the attractions were allocated by order—first come first served—rather than just money. Now, different attractions had different time costs instead of money costs. If you really wanted to ride Space Mountain, you had to pay the long wait cost of standing in the line. In contrast, you often could just walk on to It’s a Small World.

For me, the golden era of attraction allocation was between 1990 and 2021, when Disney theme parks used the FastPass system. Kiosks outside the popular attractions dispensed paper tickets with a time window on them, say 10:30-11:30. When you returned to the attraction after the start time of the window, you could use the express line, which usually meant a very short wait. FastPasses were like reservations and they were limited in number.

A key element of the FastPass system was that you could get a new FastPass a few minutes after the window from when your prior FastPass opened. So if you got a 10:30 FastPass for Big Thunder Mountain, you could get another FastPass, for Space Mountain, say, at 10:35. This was the case whether you used your FastPass for Big Thunder Mountain or not. A second key element, not widely known, was that the window wasn’t really a window. A FastPass for 10:30-11:30 could be used any time after 10:30, all the way to the close of the park. The time was less a window and more like a threshold.

These two elements meant that careful planning was rewarded with fistfuls of FastPasses. When my friends and I visited the parks, we would organize our time to be at a kiosk to get a new FastPass as soon as we could, and accumulate them. We never had the chance to get a new one without taking advantage of the opportunity. I also knew that occasionally the kiosks would malfunction. These were gold mines. When a kiosk malfunctioned, Disney employees would hand out FastPasses without checking if you were eligible to get one. (How could they, really?) In this way, you could get a “free” FastPass, one that didn’t change your eligibility for a new one.

The FastPass system was more or less egalitarian. You couldn’t buy faster access to attractions. Having said that, yes, there was a subsidy for planners and a tax on the disorganized. As a planner, I loved this system.3 Again, all allocation systems lead to winners and losers. This has to be true for any scarce resource.

Times have changed once again.

The new system Disney uses is complicated, with multiple options including Lightning Lane Multi Pass, Lightning Lane Single Pass, and Lightning Lane Premier Pass. For a price, you can skip the line, as in the Fast Pass system, for one or more attractions. The prices depend on when you visit, but can be as high as $449 per person. Note that this price is in addition to the price of admission, which might be as high as around $200 for a single day. A family of four that doesn’t want to wait in lines might conceivably pay north of $2,500 for their day at the park.

In sum, different regimes created different winners and losers:

Tickets – Money cost and time cost. Some benefit for wealthier people.

All-you-can-ride – Time cost only. Egalitarian, holding aside the fact that getting into the park in the first place is expensive.

Fast Pass – Time cost only.4 Favors planners and people with knowledge of the system.

Lightning Lane – Big money cost and small time cost favors those willing and able to spend more money.

Let’s turn from Disney back to psychology.

I’ve discussed this before, so this is a reminder of a study I conducted with my former student Peter DeScioli.

We were interested in people’s perceptions of the “fairness” of different allocation regimes. To get at this, we developed a simple laboratory study in which subjects were randomly assigned to one of two tasks. One task was easier than the other, so I’ll just refer to them as the Easy and Hard tasks.

Money to compensate people for doing the tasks was allocated in two possible ways. The person doing the Hard task could get more than the person doing the Easy task—they had to work harder, after all. Alternatively, the money could be divided evenly—the tasks were assigned randomly, after all.

We asked subjects how they felt about the two ways to allocate the rewards before they were told whether they would do the Easy or the Hard task. How fair and moral was each one? We asked them the very same questions five minutes later, after the roles were assigned and the tasks were complete.

Imagine, if you will, a rational, principled subject whose opinions about allocation regimes isn’t biased by which regime happens to benefit them. Their answers will be the same both times.

Now imagine a fallible, biased human being, whose mind is designed to try to suppress allocation regimes that are disadvantageous to themselves. Before they know whether they’ll do the Hard or Easy task, they’ll be indifferent about how the tasks are rewarded. But after they have done the task, the person who did the Hard task might decide it’s more “fair” if hard work is rewarded with more money. Symmetrically for the person assigned the easier task.

That’s what we found: people’s views about which system was fair and moral changed, in mere minutes, as their position changed.

The real world provides many such cases. Two anecdotes from my experience illustrate the point.

When I was in college, nearly everyone was politically liberal. There was one lone dissenting republican. After graduation, she found herself in the Bay Area, doing well but faced with the brutal reality of the sky-high prices of the rental market. Suddenly, she found herself in favor of rent control.

When I was in graduate school—also a bastion of political liberalism—I met one student who was a socialist, with big ideas about the importance of huge wealth transfers from the rich to the poor. He would go on to get a job at the Columbia Business School. Next time I met him, he patiently explained to me that taxation was actually theft.5

Indeed.

Theme parks aren’t the only place where allocation regimes change. It used to be that airline tickets came with a baggage allowance that was sufficient for most people. Now, people who want to bring more must pay for the privilege. Generally, various industries have moved toward such disaggregation, separating goods and services that used to be sold together. And while it’s true that many people dislike the fact that they have to pay more for their luggage, plenty of other people are pleased that they don’t have to pay for a luggage allowance they don’t use.6 Similarly, lots of people like being able to pay for a better seat, valuing the marginally improved experience more than the cost.

These debates get heated. In some cases, people judge allocation systems harshly, expressing outrage.7

Now, some moral intuitions are more or less universal. People everywhere condemn cheaters and murderers.

But there is also moral variation. If we suppose that people use moral arguments to try to persuade others that the rules that benefit themselves should be in place, then it makes sense that some people can look at an allocation regime and judge it “fair” while others judge it to be “unfair,” sometimes in the extreme.

If you are ever wondering if one of your own moral judgments about what’s fair or not is really just your self interest in disguise, change “fair” to “good for me and people like me.” If you’re a graduate student, allocating goods based on prices is “unfair” in the sense that it’s not good for you. You’re poor. If you’re a professor in a business school, taking from the rich to give to the poor is “unfair” in the sense that it’s bad for you. You’re rich.

When the rules affect everyone equally—everyone must start at the same time in a footrace—then it’s genuinely unfair if someone takes a head start. The notion of fairness applies best to rules that affect everyone the same.

In that Reddit post that I linked to above, user plurmeow writes, “I miss FastPass. Capitalism is so awful.”

I find this interesting because Disney theme parks themselves are in many ways the quintessential result of capitalism. That user likes capitalism well enough when it produces a theme park that they can enjoy because they can afford it—allowing them to have an experience that so many other people can’t—but hates capitalism when the prices rise to the point that others can purchase an advantage they can’t afford.8

This explains the point in the tagline. Lots of people love the extra pile of fries they can get for a small cost because they can afford that small cost. The premium for skipping the lines, however, is “unfair” because most people cannot afford it. To make matters worse, the people in the standby queue have to watch others walk past them. The unfairness is visible—and makes guests worse off—in a way that the upsized fries do not.

Again, it’s worth noting that these on-the-ground important differences in terms of who wins and who loses under different regimes lead to predictable differences in policy preferences. It’s equally important to note that the discussion about many of these differences is often framed in moral language.

Simplifying tremendously, modern political conflicts have been and still are about allocation regimes, writ large. One way to think about the cold war is through this lens. On one side is capitalism, a system in which stuff is allocated with prices. On the other side is a system in which stuff is allocated by a central planner, “to each according to their needs.” Other political debates have a similar flavor.

And, of course, these debates often take on a distinctly moral tone. Ayn Rand, in The Fountainhead, gives voice to this idea, that the pursuit of profit is moral and virtuous and free riding is the opposite. Participation in industrial creation is valorized. Compare this view to the message of, say, Ishmael by Daniel Quinn, in which capitalists are “takers,” to be condemned rather than praised. Rand’s readers see themselves as productive creators; Quinn’s readers see themselves and others as victimized by industrial civilization. Each philosophy flatters its adherents.

There’s a reason one never meets a poor libertarian, as the saying goes.

It’s not surprising that people favor regimes that favor them and people like them. Even more interesting is that people’s favoritism is delightfully nuanced. Some people want just enough capitalism to get them (but not others) in the gates of the park, but just enough socialism once inside others can’t buy their way in front of them.

Are we all Goldilocks Capitalists?

Most people who spend time at an all-inclusive resort on the Caribbean coast enjoy the warmth of the sun, the soothing sounds of surf, and the unlimited bounty of the buffet.

I enjoyed those things, but I also enjoyed watching the principles of economics in action.

As most readers will know, at an all-inclusive resort, you pay one price that entitles you to a room, amenities on the property, and unlimited food and drink, including alcohol. The details vary from one resort to another, but the basic idea is that you pay up front and everything else is free or, to be more precise, there is no additional (marginal, as economists say) cost. You paid a price to get access to the lattes at Mike’s Café, but the price of each latte is zero. The result is a fascinating mix of the answer to perhaps the most basic question in economics: how are goods and services to be allocated? This simple question lies at the heart of some of the world’s greatest conflicts.

There always has been, and there always will be, scarcity. Even if the AI and robotics revolutions drive prices down, only one person can live in the house on the cape and only one person can marry the fairest of them all. You just can’t get around scarcity. There is only so much stuff and some stuff is better than other stuff.

One way to address scarcity is to use prices. This is how my partner and I gained access to our five nights at the resort. The resort posts a price. The people willing (and able) to pay that price stay there. Vacationers win because they can choose the resort that fits their preferences and budgets. Resorts win because they choose the price that maximizes their profits. Price is a means familiar to Adam Smith and most humans over the last 3,000 years or so.

According to theory, under the price allocation regime, people on both sides of the transaction are better off. Buyers win because they only buy goods and services at prices lower than their value of the good. Sellers win because they only sell goods and services at prices higher than their cost. So, every transaction makes both parties better off.2

The price system is often viewed as efficient, getting goods to the people who value them the most.3

But once you are at the resort, you are in a different market. Let’s start at the end of the day, with dinner. Our property had a buffet large enough to accommodate pretty much everyone staying on property and five sit-down restaurants, believed by guests to be better than the buffet.

In Philadelphia, like most places in America, if one restaurant is in greater demand—perhaps because it wins a Michelin star—it sets a higher price, and only people who want a nicer dinner will pay it. On the resort, no one is paying any price to eat dinner, so prices don’t determine who eats at the fanciest place. (This is not exactly true, a point to which I’ll return.) So how do tables get allocated at the restaurants?

It varies from one property to the next, but at ours, the answer was basically a lottery. Reservations for the fine dining restaurants all open at 10am for that evening. You can watch people in the large buffet dining room staring at their devices, open to the reservation page for the resort’s web app, as 10am approaches. When the hour arrives, everyone clicks the “reserve now” button and grabs whatever times are still available at the restaurant in a frenzy of clicking and tapping. The lucky ones snagged a reservation at one of the nice places at a reasonable hour. The less lucky ones either got an early/late reservation or, worse, were consigned to the buffet.

Here, where allocation is random, the resort is more or less neutral—it doesn’t make any additional money when people eat at the restaurants4—and on the consumer side, the winners are those with the best luck.

Now let’s talk about beach chairs.

Perhaps a couple hundred beach chairs are set around the various pool areas. These are communal resources. They don’t have prices so they are allocated in the time-honored tradition for places to sit near water but not going into it: first-come, first-served. As something of an aside, it’s true that this way of allocating beach chairs is common, not only at all-inclusive resorts. But prices aren’t unheard of. Last year when I was in Greece, the first three rows of beach chairs had one price—I think it was about ten euros for the day—one coffee included!—and chairs in the back had a lower price, about half that. Where there are property rights, there can be prices.

In any case, the first-come first-served works fine for allocating some kinds of goods, such as the tennis courts down the block from me. And it’s better than the randomness of dinner reservations.

Now, in the case of the dinner reservations, there is no relationship between how much one would pay for that reservation and the chance of getting one. With beach chairs, though, there is something like a time price. You might say that a guest’s willingness to get up in the morning to claim a chair reflects their value of that chair. Early-risers are willing to pay the opportunity cost of sleeping in. So, there is some efficiency in the system, more than the randomness of the dinner reservations. The chairs are going to those who value them enough to pay the sleep cost, let’s call it.

Or does it? Resorts like ours have a rule designed to prevent people from claiming a chair for later. The rule is that if you are gone from your poolside beach chair for more than 30 minutes, the staff will remove your belongings to the lost and found.

Thoroughly predictably, this rule was never enforced, probably because the staff have no particular incentive to do so. Not only do they not benefit from enforcing the rule, they could alienate people who might otherwise tip them.

After spending one day at the resort, the lack of enforcement becomes perfectly obvious, as dozens of towels adorn far more chairs than there are swimmers in the pool. The permissiveness of the staff becomes common knowledge.

So what should we predict? Well, there is a clear, obvious, and implicit norm that leaving a towel and even a small trace of your stuff on a chair is a claim to a property right over it. So early in the morning, guests drop towels and a book or what have you on nearly every beach chair by the pool, starting with the ones with some shade.

From this we see the winners are not just early risers, who don’t mind the opportunity cost of getting up early, but rule breakers, who don’t mind flouting the rules. (Needless to say, we were not among these miscreants.) Symmetrically, the losers under this regime are rule-followers, who get the shaft and have to stand around awkwardly as they wait for people to call it a day and leave the pool area.

Meanwhile, the resort neither wins nor loses under the regime. The resort and the staff have no stake in which guests are on which chairs.

Now let’s turn to an area in which the interests of the guests and the resort conflict: the swim-up bar.

It’s hard to beat drinking at a swim-up bar, the novelty of having a pina colada while in waist-deep water. It’s peak vacation time.

Unsurprisingly, while there are many bars around the property, the swim-up bar is among the most popular. If people were paying per drink, management would have a strong incentive to ensure that the swim-up bar was well staffed, as each drink represents more profit. But people are drinking at zero marginal cost. So management has no incentive to keep wait times down at the bar. In fact, they have an incentive to delay the process as much as is tolerable, since every drink is a cost. The bartenders, likewise, receiving no tips—people tend not to tip on free drinks at the resorts, especially at the swim-up bar, since no one has their phone or wallet on them—have no particular incentive to be speedy. Drinks are allocated again on a first-come, first-served basis.

The thoroughly predictable result of all of this is long waits at the swim-up bar. Drinks are allocated not by how much different people might be willing to pay, but by who is able to capture the attention of a bartender. Often this is by waiting at the bar, but there are other means, which I leave to the reader’s imagination.

The resort wins this battle because it serves fewer drinks and reduces its costs. Impatient people who get out of the pool and just go to the damn lobby bar lose this skirmish.

What about service elsewhere?

At regular resorts—a la carte rather than all-inclusive—servers often continuously filter through the pool and beach area taking orders. Their earnings depend on how much food and drink they sell because—in the U.S., at least—they are tipped. They are often right at the edge between attentive and pestering, depending on the property.

At our all-inclusive, staff were always polite, but… scarce. This is not to cast aspersions on the people themselves. It is to say there were few people tending to a lot of guests and they were somewhat aloof.

Some services, however, are not included in the all-inclusive resort price.

For example, you can pay to have a massage in the one palapa on the beach. The price was about the same as one night at the property. Further, the people selling this service were on commission, earning more money with each sale. By the time we left the resort, the woman who sold us our massages was easily, far and away, the person to whom we had spoken the most. I half expected her to invite us home for dinner. She asked us when our last day was on three separate occasions. No one else on the staff seemed to care.5

Unsurprisingly, far from being aloof, these staff members were quick to pull aside passing guests and circulate among the people reading their books by the pool. I wouldn’t go so far as to say intrusive. Let’s say diligent.

A snorkel excursion to the reef was also not included in the price. I will leave it to the reader to guess how the people selling those services acted.

The thoroughly predictable result of various incentive schemes appeared everywhere.

Those sit-down restaurants? It will probably not surprise you to learn that the diner’s choices were circumscribed. No, you cannot order a soup and a salad. You may choose one item from each of the three categories: starter, entrée, and dessert.

You may have wine with dinner, yes. But we will not offer it to you; you have to (awkwardly) ask. But here is a list of expensive bottles of wine you can purchase. Take your time, let me know if you have any questions!

That steak you ordered? Yes, it’s good enough that you won’t leave a bad review for the resort as a whole but also small enough that you aren’t literally eating into their profits. After all, the restaurant isn’t going to live or die because of google reviews.

There were also some peculiarities I didn’t completely understand.

As we walked by the sit-down restaurants in the evening, they seemed to be operating well below capacity even though we were unable to get a reservation. I suppose this means that serving guests at those restaurants is more expensive. It seems like a bad look, though, and it’s hard to imagine their marginal costs were all that large, but who knows?

One way to think about this, if one wants to get a bit technical, is that the restaurants are basically living in a command economy, as one finds in communist systems. Management (the government) sets the price, in this case zero. Consumers respond by demanding the good at that price, leading to queues and rationing. In this case, you can’t really queue, so you get rationing-by-app. The restaurant doesn’t respond to the surplus demand by raising prices or increasing supply because neither is under its control.

On that note, of queuing, people were held for some time outside the buffet even when there was clearly plenty of seating available in the restaurant. It’s not clear what the resort’s incentives are in this case. If you have a thought, feel free to comment. It didn’t seem to me that anyone was better or worse off, except the guests who had to wait a bit more.

From all of this, we can take several lessons.

First, there is no perfect solution to the inevitable problem of allocating scarce resources. Consider some of the ways humans allocate valuable resources:

1. Prices. This way leads to efficiency. Many people like this one, but especially those with a lot of money, the thing used to get stuff.6

2. First-come first-served. This way gives you some efficiency. Just think about the people who queue for iPhones and Duke basketball tickets. They wouldn’t be there if they didn’t want those goods more than others.

3. Need. Emergency rooms generally work this way.

4. Merit. Many people think this is a good way to allocate things like jobs and awards.

5. Force. Whoever can take it gets it. Favored by people like Genghis Khan and Alexander the Great. Less popular with those without weapons, defenses, or power.

6. Mutual preferences. See: Tinder.

7. Identity. Some stuff is allocated, or allocated preferentially, based on identity characteristics such as sex, ethnicity, religion, and so on. Nepotism falls in this category, and was favored by people like Henry VIII and Joffrey Baratheon. Not really efficient, but this kind of system helps to solve a coordination problem.

8. Lotteries. Beach chairs, yes, but also places in certain schools.

There are other ways as well. All have their advantages and all have their drawbacks.

Maybe these systems are better for some goods and services. Maybe lotteries are a good way to allocate some things in life, though I struggle to come up with good examples. Certainly merit seems like a good way to allocate prizes and positions of responsibility. The point is, getting rid of prices solves some problems but invites others in. There is just no way around scarcity.

Second, all-inclusive resorts are little laboratories where you can look at different ways to allocate goods and services. In particular, these resorts illustrate to people who are accustomed to prices what happens when you don’t use prices to allocate stuff.

On the one hand, people like it. People don’t like prices—who wants to pay for stuff? But once prices are gone, something has to determine how things are allocated. We see what it does to dinner reservations, but let’s expand out. Let’s say that health care is a right and that as a society we don’t want to make people have to pay for being healthy. Ok, but then how can we figure out how much health care to make and who gets it? Could healthcare come to resemble the swim-up bar, with long wait times and a mediocre product? Without prices, we invite a zoo of allocation systems to take their place, as indicated above.

Related, as economists have noted, figuring out how much stuff people want is hard. Those empty seats at the restaurants? Could that be because the resort literally doesn’t know what people want? This is essentially a shortage economy: there’s excess demand and at the exact same time excess capacity because prices aren’t adjusting to coordinate them. The antidote to this problem is, of course, prices, which carry information. Without a price signal, as economists say, the market can’t clear. That is, following people such as Adam Smith and Friedrich Hayek, sellers raise prices when there is excess demand and lower prices when there is excess supply, pushing the price to the point at which supply equals demand. In theory. Prices then become signals of the value of the good or service.

Third, individuals and groups respond to incentives in thoroughly predictable ways. The firm limits drink consumption. Staff work hard when they are compensated on the margin. Guests break rules to gain positional goods.

For my part, I like economic efficiency—the distribution of goods and services to those who most value those goods and services—so I, like Esther Dyson, very much like the price system and very much don’t like when prices are distorted or absent. There are exceptions, such as the emergency room and jobs, but for most things people want to access, prices are the right way to pair demand with supply.

To say that behavior is motivated by emotion captures a wide range of human tendencies.1 People can be observed:

• Keeping a good feeling going, as when we continue to eat the open bag of Doritos in front of us.

• Pursuing an imagined future good feeling, as when someone grinds to make partner at a law firm or achieve tenure at a university. Capturing that feeling, of course, often requires misery in the interim—an interim that, in extreme cases, might constitute the bulk of someone’s life. But even someone miserable for most of their life is still, in this sense, pursuing a good feeling.

• Fleeing or avoiding a bad feeling, as when we procrastinate on the expense report, despite increasingly urgent reminders from accounting, because filling it out is so mind-numbing.

• Accepting bad feelings to keep worse ones at bay, as when someone stays in an unhappy marriage because it’s preferable to loneliness, a sense of failure, or the worry/guilt of screwing up one’s children.2

Feelings, it might be said, are a bit like gravity: sooner or later, they exert their pull on the human animal. Just as breasts drop, spines compress, and shoulders stoop, people yield to varying degrees to their emotions.

Motivational States

How someone responds to the claim that “humans are unique in their ability to feel emotions” usually tells you more about them than the claim. Some will agree wholeheartedly; others will rush to defend the emotional lives of whales and dogs.

While some animals do indeed show rudimentary forms of emotion, many are motivated by what has traditionally been called “instinct.” The male rhino appears to climb onto the female with the same enthusiasm I reserve for doing the dishes; the female displays about as much emotionality as my wife waiting for the train. The whole proceeding has the feel of a chore—a task—for which there is no need for melodrama.

At any rate, the claim about humans’ emotional uniqueness loses all its bite—and becomes so obvious as to be barely worth saying—once we frame it as a matter of degree rather than kind. As William James observed, humans don’t lack the instincts found in other animals; we have more of them.3 Humans experience a wider range of emotions than other animals, just as we use a wider range of communicative symbols. The apparent counterexamples—our domesticates (cats, dogs) and close evolutionary cousins (primates)—only reinforce this point.4

All this said, the difference between an instinct and an emotion doesn’t have much functional significance. Whatever compels the male rhino to mount the female is the same, in effect, as the emotion that compels me to avoid filing an expense report. Is it important that the rhino’s is labeled an instinct and mine an emotion? Not really.

So, I’m using “emotion” and “feeling” as convenient shorthand for internal states that move us to act, which have variously been called instincts, drives, impulses, sensations, and bodily urges. By grouping these under a single heading—motivational states—we arrive at the pleasant tautology that humans are motivated by motivational states, and can safely conclude that we haven’t explained much so far.

Proximate Targets

Feelings compel us to act in ways that, over evolutionary time, tended to correlate with adaptive outcomes. We’ll return to how that relationship has frayed in the modern world, but the important point for now is that many of the actions our emotions motivate are not adaptive in and of themselves.5 Emotions steer us toward proximate targets, not ultimate evolutionary ends.

The classic example is sex. Humans enjoy sex because, in the past, it reliably correlated with reproductive success. But sex is not the same as reproductive success, a distinction that has become obvious in a modern world replete with contraceptives and IVF. Indeed, for many people I know—including my wife and me—sex is a messy and unreliable method for producing a child. Just let someone in a lab coat do it.

In the ancestral world, the difference between sex and reproductive success was largely semantic. In the modern world, which tends to widen the gap between proximate action and ultimate outcome, it isn’t.

Still, why didn’t evolution simply motivate us to achieve the ultimate goal of reproductive success? Why did it aim at proxies instead? Because aiming at ultimate goals is nearly impossible given the sheer number of variables in play, especially for a highly flexible species like ours. So, instead of trying to motivate us toward maximizing our genetic contribution to future generations, evolution motivates things like seeking sex, avoiding social exclusion, pursuing status, eating calorie-dense food, and protecting our children because these are simple targets compared to an abstract calculus of inclusive fitness.6

Now let’s return to the effects of the modern world. Just as sex no longer correlates with reproductive success the way it once did—making the emotion that motivates sex, lust, less reliable—the same is true of emotions more generally. From both a practical and evolutionary perspective, nearly every emotion is now more suspect than it once was. Take anger. In the ancestral past, anger was often useful for raising others’ estimation of you; it functioned as a “take me seriously” signal. Applied to strangers in the modern world, however, most of whom we’ll never see again, anger makes far less sense. In fact, it’s usually counterproductive.7

In light of this, some might be tempted to call indulging in anger “maladaptive.” But anyone familiar with evolution would move quickly past that term, lest it obscure the adaptive history of the emotion. Right?

In any case, the point is that lust, anger, and many emotions across many different contexts are less reliable than they once were. The ancient—and for a long time only—rule of emotion, act on it when you feel it, no longer fits the world you and I inhabit.

Aging into Heuristics

So far we’ve established:

1. Human behavior is motivated by emotion. (Hunger motivates me to eat more Doritos.)

2. Emotions steer us toward proximate goals that, over evolutionary time, tended to produce adaptive outcomes. (Lust motivates sex; sex once reliably led to reproductive success.)

3. In the modern world, that ancient link is often broken. (Sex no longer reliably leads to reproductive success, yet the motivation remains. (Does this lend to much of our activity a degree of pointlessness? Does it cause us to occasionally resemble mice on a wheel? I digress.))

4. Every emotion is potentially vulnerable to this disruption. (Anger, for example, is more likely to be counterproductive in a world full of strangers and police.)

5. As a result, modern humans must frequently interrogate their emotions rather than simply obey them. (If nothing else, this is kind of a slog.)

I wonder if a similar process begins to take over the behavior of an aging person: doing the type of thing that used to correlate with success and caring less about what success ultimately means.

For example, I was cleaning the house the other day, which required hauling Halloween costumes to the attic (get off my back). As I set the box down beside the others, I noticed something odd: the Halloween one wasn’t labeled.

Should I label it?

One part of me said no—after all, I’ll only need this box once a year, and when I do, it will be perfectly obvious which one it is. The sides are transparent, and Halloween costumes are hard to miss. But another part of me said yes—just label it. It will take two minutes, and it’s the right kind of thing to do. That is, the action meets the standards of consistency, thoroughness, and similar virtues that pay off over time.

Rephrased slightly, the first part of me is asking: What am I ultimately trying to achieve, and which sequence of actions best achieves it? The second part is saying: Don’t worry about all that. Just adopt a rule that tends to correlate with success and follow it every time.

As it turns out, this distinction parallels the divide in moral philosophy between the utilitarianism of John Stuart Mill—do the thing that produces the best outcome—and the deontology of Immanuel Kant—adopt a rule (e.g., tell the truth) and follow it regardless of the immediate consequences. Weighing the costs and benefits of labeling the box is a utilitarian move. Labeling it simply because it’s a box, and boxes get labels, is a Kantian move.

The utilitarian approach, which I favored more as a young buck, has the advantage of potentially saving time. The deontological approach, which I seem to favor as I get older, has the advantage of reducing cognitive load: once a rule is adopted, no further decision is required. I simply execute.8

Why might the deontological approach become more preferable with age? Because:

1. Responsibilities accumulate. As decision volume increases, simple rules become more appealing.

2. Perception sharpens. With experience, people become better at detecting nuance. Ironically, this raises the cost of case-by-case reasoning, since more variables must be weighed each time. Heuristics—simple rules or mental shortcuts—allow that complexity to be bypassed except when it truly matters.

3. Failure leaves a residue. Even rare failures of the utilitarian method tend to lodge in the bones, nudging people toward conservatism—an orientation that favors stable rules over clever (but ultimately risky) optimization.

Of course, given the complexity of life on this planet, and the much higher cost of failure than success, evolution has settled on the same solution: adopt the rule that tends to work and follow it.

In other words: just label the damn box.

At this point, readers might be wondering: has psychology gotten anything right?

In mulling this question, we came to an unsurprising conclusion. On the harder side of the field, in subdisciplines such as neuroscience, animal learning, perception, and cognitive science, psychology has produced genuinely impressive work. On the softer end, in areas such as social and clinical psychology, not so much.

Why is this unsurprising? As others have noted—in Paul Meehl’s case, more than 50 years ago—claims in the hard sciences are often specific and falsifiable, whereas claims on the soft side often are not. If you say a certain neural circuit does X, or that an organism will learn Y under Z conditions, someone can rerun your experiment and find out. False claims don’t survive long in that environment, which means fewer of them are made in the first place.

As the discipline softens, however, constructs become vaguer and predictions looser. The harder a claim is to test, the harder it is to refute—and the more room there is, sometimes even incentive, to make something up. This need not be nefarious, and usually isn’t.

The questions clinical psychology wrestles with are no less important—or discoverable—than those studied by vision scientists. The difference is that the vision scientist has sharper tools, tighter measurements, and clearer feedback. Figuring out why humans think, feel, and behave as they do is more complicated than mapping a visual cortex. Indeed, psychology’s mission may prove even more difficult than explaining the physical universe.

This is why the evolutionary approach is so crucial. Because the subject matter is so difficult, we ought to use every tool at our disposal. Asking some of the hardest questions in science while refusing to use the strongest available theoretical tools is negligent. Unsurprisingly, it has produced—and will continue to produce—many bad ideas.

Before proceeding to some of the good ideas, a caveat: just because the proposals below have withstood falsification so far does not mean they should be taken as Truth. To a first approximation, all models are wrong, or at least incomplete. Science is a process that, at its best, replaces one wrong proposal with one that’s a little less wrong. The “hits,” as we call them, have survived until now and we are optimistic about them. But it’s not as if they are established laws of nature. Still, it’s worth celebrating some gains that we think will endure in some form.

Laws of Learning—RK

Including the laws of learning might seem a bit odd to those with a deep background in evolutionary psychology. After all, much of the foundational work in the field was developed in no small part as a rebuttal to the idea that behavior could be explained with domain general laws of learning, i.e., that the human mind is initially a blank slate that is written on by what the individual perceives through their senses.

Fair enough.

Still, this work represents excellent research that has stood the test of time. Behaviorist principles are still used to train animals, including humans in the form of gamification. Just think of how much less walking there would be without the satisfaction of reaching the 8,000 step goal so many people have adopted. And I personally can now get by in Spanish-speaking countries thanks to the gamification of Duolingo. (Look for Diana Fleischman’s upcoming book on how you can use the principles of learning to train your romantic partner.)

Gamification is downstream of early research on learning. On that note, I was fortunate to overlap with Bob Rescorla while I was at Penn, so perhaps I am biased, but consider the Rescorla-Wagner model. Put roughly, this model formalized the idea that animals learn from being “surprised.” Consider Pavlov’s famous dogs. Oh, there’s food right after that bell rings? I did not know that. I will update my belief about what that bell foretells.

But the model wasn’t just that. It was explicit, saying what the relevant variables were, and those variables had units. When you propose an explicit mathematical model, not only are you bravely exposing your model to being wrong, but you are exposing your model to be wrong by a specific amount. The models we discussed in Parts I and II aren’t explicit and quantitative in this way. Some of them are sufficiently vague that it’s unclear if any observations would falsify them. Others just make directional (ordinal) predictions: you will feel stronger, not this much stronger, if you pose in this way.

That’s not to say that the Rescorla-Wagner model is perfect. It’s been extended and refined over the years, but the general idea has been robust. The key idea is that animals associate one thing with another—bells and food—and that these associations get changed when the animal is surprised (a “prediction error”). That change in associations is “learning.” The corollary—that if an animal isn’t “surprised” then it won’t learn anything—also remains more or less intact. The scientific success of the Rescorla-Wagner model is due in no small part to the fact that it was quantitative and explicit, allowing it to be tested.

Now, to be fair, one could argue that modeling how fast a dog learns to salivate to a bell is easier than modeling how a child learns language or how a person gains confidence. Still, the model was a signal achievement and maybe even illustrates what the social sciences could look like down the road, when scholars have sufficient confidence to put units and values to their models.

Ecological Rationality—JZ

Beginning in the 1970s, Daniel Kahneman and Amos Tversky ran a series of clever experiments suggesting that people do not reason the way a purely rational agent would according to the rules of formal logic and probability theory. For example, humans neglect base rates when judging probabilities, are loss averse when weighing gains and losses, anchor on arbitrary numbers when making estimates, and prefer certainty over expected value when choosing between risky options. Over time, the list of ways humans actually reason—compared to how economists and other academics assumed they should reason—grew long. It is captured in this awesome codex of cognitive biases:

Kahneman and Tversky’s work toppled a prevailing view of human nature at the time: Homo economicus, the cool, calculating agent who maximizes subjective expected utility. Compared to Homo economicus, Homo sapiens began to look pretty stupid. The language of “bias” and “distortion” took hold and, in many quarters, hardened into a broader narrative that humans are fundamentally irrational.

That’s where Gerd Gigerenzer and others entered the picture.

Gigerenzer and his collaborators argued that humans only seemed irrational because the wrong model of rationality was being applied. As might seem obvious to anyone who regularly reads this Substack, humans were not designed to solve abstract probability puzzles in a lab. They were designed to make decisions under uncertainty in real environments—environments with time pressure, incomplete information, and recurring structure.

When problems are presented in ecologically meaningful formats—for example, in frequencies (2 in 10) rather than percentages (20%)—people’s reasoning often improves dramatically. Simple rules like “recognition equals importance” or “imitate the majority” are not necessarily cognitive flaws. In many environments, they are fast, frugal, and remarkably effective. As I argued in Moving Beyond System 1 and 2, what looks like bias in the lab is—or at least was—adaptive in the wild.

That’s why, if we insist on calling these heuristics “biases” or “distortions,” we should put those words in quotes. The real distortion lies with anyone who presumes the mind should work differently than it does. The irony, of course, is that our brains’ ecological rationality has been so successful that some of us can sit around inventing artificial problems to confound it, then mistake the confusion for evidence of design failure.

In short, humans are ecologically rational—not economically, formally, or academically so. Anyone who evaluates a mind designed for one ecology using the standards of another is bound to mistake design for defect. Correcting this mistake has been one of psychology’s real achievements, although I daresay the average psychologist—let alone the average therapist—remains unaware.

Family Stuff - RK

My advisor team in graduate school was a married couple—a cognitive psychologist (Leda Cosmides) and a biological anthropologist (John Tooby). For them, science was both a career and a family affair.

Like many advisors, they assigned me and the other graduate students copious reading, especially during my first year. The classic ethnologies—e.g., Nisa and The Fierce People—illustrated how central family and kinship were to, well, pretty much every culture anthropologists studied. I was also introduced around that time to the Human Relations Area Files, a vast corpus of information about different cultures accumulated by anthropologists, and Don Brown, author of Human Universals. These three sources persuaded me of the centrality of family to human social behavior around the world and very possibly since the dawn of the species.

Then again, family and kinship were oddly absent from the social psychology sources I was reading. There were discussions of “close relationships” and “helping behavior,” but precious little about the family. There was a little bit in clinical psychology, but if you picked up a social psychology textbook—putatively about human social behavior—you wouldn’t know that cultures primarily organize their social lives around kinship.

In 1988, Martin Daly and Margo Wilson published the book Homicide. Drawing on ideas from evolutionary biology and behavioral ecology, they proposed that there is a big difference between raising a child who was one’s own and a stepchild. Using data available at the time, they showed that there were much greater dangers of abuse from a stepparent than from a biological parent. This finding is consistent with the theory of kin selection and, of course, the anthropological record. While there has been some controversy about the size of this effect, the finding has withstood the test of time.

Related, as it were, I overlapped in graduate school with Deb Lieberman, who would go on to extend an idea associated with Edward Westermarck. Westermarck proposed that people who are raised in close physical proximity to one another naturally develop sexual aversion to one another. The idea was rooted in the biological cost of mating with a close biological relative. Westermarck’s work was supported by later research by Arthur Wolf, who studied “sim-pua” in Taiwan, the practice of bringing a young girl into the household of the boy to whom she is to be wed. Wolf found that such marriages are more likely to fail than marriages in which spouses meet one another later in life. A related finding held in Israeli Kibbutzim.

Deb Lieberman, building on these ideas, proposed that humans, as they develop, compute a “kinship estimate” from various cues, not the least of which is co-residence. Seeing your parent provide aid to another little person might be a good cue that you are related. In any case, Lieberman added to the body of evidence illustrating that humans naturally develop a sexual aversion for those they compute to be closely related.1

Both of these findings have been the subject of a certain amount of debate, but my view is that they have held up reasonably well and provided the basis for additional avenues of work. Our guess as to why is that they are grounded in good theory, make clear testable predictions, and align with neighboring disciplines.

The Unconscious—JZ

For much of Western intellectual history, reason was treated as the defining human faculty. The mind was assumed to be unified, coherent, logical, and largely transparent to introspection—a conscious agent steering its own behavior.2

Then came Sigmund Freud.

Freud was wrong about many things, but he was right that much of what drives human behavior occurs outside conscious awareness. His theories of repression, defense mechanisms, and hidden fantasies were often speculative and difficult to test, but the central insight—that the mind is not fully transparent to itself—has endured.

Modern cognitive science has reinforced and clarified Freud’s insight. For example, we initiate movements before we become aware of deciding to move. We often generate explanations for our choices after the fact, as when Aditya justified his gorging on chocolate cake as “carbo-loading” before a run. In classic split-brain experiments, actions initiated by one hemisphere were followed by confident verbal explanations from the other, even when the speaking hemisphere had no access to the true cause.

Partly as a result of this research, new metaphors for consciousness have taken hold. The conscious self is no longer cast as captain of the ship but, in Jonathan Haidt’s formulation, as a rider atop an elephant—capable of guidance, but only within limits. Or, as Rob put it in Why Everyone (Else) Is a Hypocrite, the conscious mind functions more like a Press Secretary, crafting narratives on behalf of the real decision-maker: the unconscious President.

In the end, the modern view of the unconscious has come a long way since Freud, but it owes its current clarity to his initial insight. Consciousness now looks less like a sovereign ruler and more like a bandwidth-limited interface riding atop systems that are older, faster, and more opaque.

Just as with rationality, psychology took a circuitous route, but seems to have gotten this right in the end.

Signal Detection Theory—RK

England won the Battle of Britain in no small part due to radar, which gave the English critical information about incoming German attacks so that the scarce resources of the Royal Air Force could be allocated to maximum effect. Radar operators could make two kinds of mistakes: 1) reporting that a signal was an incoming attack when in fact it was really just noise, and 2) failing to report a signal that was actually an attack. The first kind of mistake wasted resources; the second kind of mistake risked leaving German bombers unopposed. Thinking carefully about this problem—the costs of different kinds of errors—led to Signal Detection Theory, a powerful model for decision making.

After the war, researchers like John Swets, Wilson Tanner, and Ted Birdsall imported the framework into the study of human perception in the 1950s. One of their key insights was that any detection problem—faced by a radar operator, person, or animal—has two distinct components. First, sensitivity: how good is the system at distinguishing signal from noise? Second, criterion: how much evidence does the observer require before saying “yes, that’s the thing I’m looking for”?

This distinction matters because the two are independent.

One’s instrument might be sensitive to small differences, or not. For example, you might have a scale that measures things to the microgram, or maybe one that just reads to the nearest kilogram. The smaller the differences a sensor can detect, the more sensitive it is.

And one can be more or less conservative in judging whether the stimulus is what you’re looking for. Take something like a smoke detector, a frequently-used example. It’s fairly good at distinguishing between smoke and non-smoke. However, because its job is to alert people of a fire, its threshold is set very low. This “low criterion threshold” means it goes off not only during a blazing inferno, but also when you burnt the toast. Moderately sensitive, very low threshold.

It’s been successful because it’s explicit, with numbers and everything. Just as in the case of learning, above, the theory can be, and has been, tested not just in kind—do the findings go in the right direction—but in degree—do the findings match quantitative predictions. Second, its success is, as with the Westermarck effect, due to the fact that it connects with deeper disciplines, in this case engineering, which grounds the work.

One way to see Signal Detection Theory’s success is that—unlike most theories used by psychologists—it’s used in real settings with concrete costs, including in medical diagnosis, eyewitness identification, quality control, and many other places a human or machine has to make a binary judgment under uncertainty. It doesn’t go through “conceptual replications” or need p-hacking to stay alive. It simply works, in the way that useful quantitative models tend to.

It’s been so successful, in fact, that it has arisen in other guises. Evolutionary psychologists imported the idea into Error Management Theory, which applied the principles of signal detection theory to the context of mating. The core idea is that when the costs of different errors are asymmetric (e.g., missing a mating opportunity vs. mistakenly inferring interest that isn’t there), natural selection should, everything else equal, shift the criterion, biasing the system toward the less costly error.

Series Conclusion

Before ending, we’d like to share some lessons we’ve learned along the way:

1. Seeking the truth is difficult—for better and for worse. It requires constant vigilance, steady open-mindedness, and routine humiliation of the ego (think you have a unique thought? Think again!). On the one hand, that’s unfortunate—many people do it poorly, and poor thinking has real consequences. On the other hand, it’s invigorating. For those who enjoy the life of the mind, reality offers unlimited access to the flow state.

2. There is more garbage than gold. Lee Jussim has persuasively argued that ~75% of Psychology Claims are False. Perhaps this is to be expected in a discipline that is both so complicated and so young (~150 years old). In any case, it means that everyone—us and other experts included—understands far less about human behavior and the brain than we often pretend. This justifies giving personal experience and intuition a more prominent seat at the table. Trust your gut, and be skeptical of what you hear.3 Appealing findings, such as those that are just counterintuitive enough, should raise extra suspicion.

3. Scientific claims must be clear and specific. Too much time is wasted on the definition game: one person says X causes Y, another says it does not, and the disagreement turns out to hinge on what “X” means. Vagueness can arise from carelessness or self-protection. If no one knows precisely what you are claiming, no one can show you are wrong—or suggest how you can become more right.

4. Claims should be quantifiable and operationalizable whenever possible. Anything that exists, exists in some quantity. Anything that is true should work when applied.

5. Claims should be consistent with other disciplines. Chemists generally don’t propose theories that violate the known laws of physics. Psychology should aspire to the same level of integration with neighboring sciences.4

6. Claims should be as simple as possible—and no simpler. Follow Occam’s Razor. Before inventing new constructs, ask whether existing ones suffice. Is transference meaningfully different from familiarity? Is trauma always more than learning? Sometimes a new word marks a new concept; other times it’s just a new word.

That’s the series! Thanks for following along. If we’ve been clear, you should now be well-equipped to give us some feedback.

Classic social-priming effects (e.g., walking slower after reading “old”) failed replication; most are statistical flukes.

2. Self Esteem

Inflating self-esteem doesn’t cause achievement. It’s the other way around.

3. Emotional Intelligence

“EQ” lacks a clear definition, consistent measurement, or strong predictive power. What looks like emotional intelligence is often ordinary social skill expressed in specific situations rather than a general trait.

4. The Neurotransmitter Model of Mental Illness

The idea that disorders like depression are caused by simple chemical imbalances (e.g., “low serotonin”), and that psychiatric medications work by correcting these imbalances, persists in both public and professional explanations (for example, try asking an AI). But the neurotransmitter model of mental illness has long since proven bankrupt. The truth is that we still don’t know what causes most mental illnesses, how they develop, or why antidepressants work—if they even do beyond placebo.

5. Birth Order Effects on Personality

The idea that firstborns, middle children, and later-born siblings have distinct personalities is widely believed. But large-scale studies find little consistent evidence for this view.

For Canada, the defeat was crushing for all the usual reasons and an additional one: they had dominated play for much of the night. Have a look at the stats:

Canada was understandably frustrated that its dominance in nearly every facet of the game failed to translate into victory—a frustration captured by center Nathan MacKinnon after the game:1

Yet it wasn’t only Team Canada players and fans who were frustrated. A friend who watched the game with me—who knows little about hockey—was bothered, too. He thought Canada deserved the win and couldn’t shake the feeling that something had gone wrong, his allegiance to the U.S. of A. notwithstanding.

I’m sure many readers know the feeling I’m describing. It’s not uncommon to watch (or play) a sport or game and come away convinced that one team or player should have won, even though the actual winner played entirely within the rules and thus won “fair and square.” Other examples include:

• A chess player outmaneuvers her opponent for most of the match before blundering her queen

• A soccer team generates triple the chances but hits the post twice and watches the opposing goalie make a string of improbable saves

• A star pitcher in the middle of a no-hitter leaves with an injury, and his team goes on to lose

• A clearly superior college basketball team plays poorly for one night and is eliminated from the tournament

• A football team exploits a technically legal but widely disliked tactic (“Tush Push,” anyone?)

What’s going on here? What’s occurring in the inner workings of human psychology to make my friend, who cares little for hockey and wanted the Americans to win, leave the party feeling slightly askew?

The Strong and Strange Emotions That Sports Elicit

Sports elicit strong emotions mainly because humans evolved to hallucinate group membership, and sports graft onto this tendency particularly well—what with their ritualized chants, distinctive jerseys, regional allegiances, and heated rivalries. But I also think sports make us feel or behave in strange ways because, although they mimic ancestral conflict, they operate very differently—sort of like how Splenda tastes like sugar but doesn’t act like sugar in the body.

I delved into one example of this in Why Is a 2-Goal Lead the Worst Lead in Hockey? My argument was that the defender’s advantage found across the animal kingdom—which one might assume would lead winning teams to protect their lead better than chance would suggest—actually backfires. One possible explanation is that the defender’s advantage evolved in response to concrete assets such as territory or resources, not the relatively abstract notion of score. As a result, winning teams often behave as though they are defending territory or resources, which causes them to sit back and, paradoxically, invite the opponent back into the contest. Hence the familiar line in football that “the only thing the prevent defense prevents is winning.”

Another consequence of mistaking sport for real conflict is this notion of “deserving to win.”

When a seemingly deserving team loses, players and viewers often experience something like a moral imbalance. The “better” team lost; the “wrong” team advanced. The reaction is not mere disappointment but, in my experience, carries a twinge of anger or indignation. Indeed, I’d be shocked if the Deserve to Win O’Meter, developed by the hockey statistics website MoneyPuck, was not intended (at least in part) to cater to that righteousness. There is some solace, when your team loses, in knowing that they shouldn’t have. That the outcome was unfair.

Note that I am not including poor officiating or outright rule-breaking. In those cases, it is easy to see why someone would be upset. The grievance is legitimate. In the examples above, however, there is no such violation. Technically, there’s nothing to complain about. A player or team won by playing the game as it was agreed upon at the outset. So why the lingering feeling when the outcome isn’t what it “should” have been? By the rules of the game, the winner is the winner—period. Why doesn’t it feel that way?

I think the answer lies in evolved cheater detection mechanisms. My claim is that people feel off-kilter when they watch a team receive the benefit of winning without paying the cost.

Humans cooperate much more than most creatures in the animal world and a key feature of our evolved psychology that supports this is that we are very good at detecting cheaters—those who enjoy a benefit but have not paid the associated cost or met the relevant requirement. Some of the foundational work in evolutionary psychology, by Leda Cosmides and her collaborators, used logical reasoning tasks to show that people have a particular knack for identifying cheaters, even in unfamiliar contexts. Because people can identify cheating and, in addition, are motivated to punish them, being cooperative can be a winning social strategy.

Sports often activate this fairness machinery. People have the sense that if one team has paid the cost or met the requirement, then they are entitled to the associated benefit: winning. Perhaps cases in which this requirement has been met but the team didn’t get the benefit—or the requirement was not met and the team did—trigger this evolved cheater-detection mechanism. (See Gerd Gigerenzer’s work on this topic for compelling results along these lines.) The Olympic gold medal game is one such example. To the average viewer, Canada worked harder, sacrificed more, or simply played better than the United States—but the United States got the reward. We thus feel—if not toward them, then toward the situation—the same way we would if we saw someone promoted over a colleague who clearly deserved it more.

But that raises an interesting question: What do we perceive as the costs of winning? It’s obviously not writing a check. Instead, costs roughly track the kinds of investments humans intuitively associate with dominance and effort. A team “deserves” to win if it plays better for most of the contest, generates more chances, faces the full strength of its opponent, is “better” as defined by prior performance, secures advantages through intention rather than randomness or luck, and plays the game in the way people expect it to be played.

In ancestral environments, cost meant risk, energy expenditure, coordination, dominance display, and prior investment in skill and preparation. In modern sports, people map those intuitions onto proxies such as territorial control, time of possession, chances generated, and visible displays of skill. When those proxies fail to align with the final outcome, our fairness alarms go off. The scoreboard declares one winner, but our intuitions declare another.

More generally, it comes across as an investment asymmetry: one team appears to have invested more, but the other collects the reward. Humans don’t like that. Indeed, if our species had not evolved a strong dislike of that dynamic, I wouldn’t be writing this and you wouldn’t be reading it.

Exceptions and Variance

A few final points. First, the “deserving to win” intuition is only one among many that sports generate. We also tend to favor underdogs, a potential consequence of humans’ leveling instinct. Thus, even though David did not “deserve” to beat Goliath—he was smaller, relatively untrained in combat, and didn’t play the game the way it was supposed to be played—we still cheer for his victory.

Put the two intuitions together and things get interesting. My sense is that an underdog who outplays the favorite but still loses often feels more deserving of victory than a favorite who outplays the underdog and loses. Another example of the extraordinary context-sensitivity of the human mind.

Finally, in the world of sports, it’s not that unusual for deserving teams to lose. Granted, this occurs more frequently in some sports (baseball and hockey) than in others (football and basketball), but generally speaking, sports exaggerate variance. Now, if misalignment between effort and reward had been more common in environments of evolutionary adaptedness, human intuitions might be more tolerant of it. But they aren’t, because it wasn’t. Fairness machinery evolved in environments where effort and reward were tightly correlated. In ancestral conflict, if you dominated 80% of a fight, you probably won.

That relationship weakens in sports, which in many ways confuse and overengineer conflict and competition (not that I don’t love them). Directing a rubber puck into a net past a heavily padded goalie while skating on blades is a far less transparent display of dominance, strength, and skill than something like boxing or mixed martial arts. I can shoot three pucks at your net and hit the post three times; you can shoot once, have it deflect off a teammate’s leg, and score. Is your victory a reliable verdict of superiority or dominance? Of who is, in some deeper sense, “better”? Not really.

Because of this, sports and other games are ecologically invalid targets not only for cheater-detection mechanisms, but also for our dominance- and status-assessment systems more generally. The “deserving to win” intuition, in specific, arises because our fairness psychology tracks effort and investment—if you pay the cost, then you get the reward—but sports don’t always work that way. Sometimes pretty wild stuff happens instead:

In sum, sports preserve the appearance of conflict while functioning differently. For one, they reward score, something no animal ever evolved to acquire or defend. For another, they weaken the link between effort and outcome, often creating two winners in the mind of the viewer: the team that won and the team that deserved to win.

Now, you might have thought that was the end of it, but I’d like two more paragraphs. Unlike with most of my articles, I’m not entirely convinced by my argument in this one. In his edits, Rob raised several points that complicated the picture more than they clarified it. For example, consider our different intuitions for hunting and gathering: we expect gathering effort to track return closely but some activities, such as hunting, naturally have a chance element, meaning that sometimes the better person wins and sometimes they don’t. For such cases, humans are more or less fine with the fact that race goes not always to the swiftest. Which are sports more like? And if we asked sports fans whether they’d want to live in a world where the team that deserved to win always did, would they? I suspect many of them prefer, at least in theory, the stochastic element.2

Finally, who is my friend miffed at? According to my argument, he should be miffed at Team USA—they are the “cheaters” in this scenario, the party that took the reward of winning without paying the cost. But he’s clearly not angry at them, nor is he angry at the sport of hockey, which failed to indicate who was better. He’s more perturbed by the situation itself. Is there something significant about that?

In the end, I suspect this article has only, you know, broken the ice on the topic. I’d be curious to know what everyone else thinks.

The first experience was at the University of Arizona, where I was fortunate to collaborate with Amnon Rapaport, a brilliant economist. He ran a series of experiments that were designed to imitate people’s decisions regarding whether or not to enter a market. So, suppose you’re considering opening a new paint-your-own pottery studio. Your decision of whether to enter this market will depend on how many other people are in that market. If there are too many, then all market entrants are worse off because there isn’t enough demand for every entrant. But if the number of entrants matches demand, then everyone is better off, including the people who (correctly) stayed out of a market that would have been too crowded if they joined.

To model this, he brought groups of 12 subjects into a laboratory where no communication among them was possible. The rules of the game were simple. There would be multiple rounds of play, and in each round, each subject could make one of two moves—either ENTER the fictitious market or STAY OUT. The payoffs were straightforward. They were highest when EXACTLY seven people entered and declined as the number of entrants became larger or smaller than seven.

The result was so surprising that Amnon and collaborators used the word “magic” in their scholarly publications reporting the results. Somehow, groups wind up with seven market entrants in most rounds and, further, which participants enter the market changes across rounds. Remember, there is no communication, so it’s unclear how they accomplish this feat. The best explanation, which might not be all that satisfying, is that everyone sort of decides how often they will enter the market—some more and some less—and the result is that you somehow get the optimal number of people.

The second experience was a conversation with another economist, Colin Camerer, during my time at CalTech. Our discussions brought me around to the view that coordination was at least as interesting as cooperation, if not more so. Colin, whose name is frequently whispered around the time the Nobel Prize is announced, persuaded me that understanding results such as Amnon’s could be very important for understanding human behavior, both in markets and other contexts.

And then came my student, Peter DeScioli, who persuaded me that the key to unlocking the mysteries of morality lay not in cooperation, as most people at the time believed, but in coordination.

One more historical note and then on to the meat of the post and the answer to the question in the title. Thomas Schelling, in The Strategy of Conflict, was interested in how people choose strategies in coordination games. The classic example is that you and a stranger get a reward if you manage—without any communication between you—to be at the same place at the same time tomorrow in New York City. Many people pick Grand Central Station at noon. The time and the location stand out. They are “focal points,” a solution that feels not only obvious, but obvious in a way that others will find it obvious. People use these sorts of obvious solutions to try to solve any number of coordination problems. They might be obvious because they are unique, famous, or related to a cultural convention.

Ok, on to eclipses.

A solar eclipse occurs when the moon passes between the earth and the sun. So the earth is in the moon’s shadow. Depending on the details, the moon might block some or all of the sun. When the moon—or the observer—is positioned just right, the moon blocks the sun completely, a condition referred to as “totality.”

A recent paper (by Miao et al.) reports an analysis having to do with how often areas rebelled over the course of centuries of Chinese history. Do provinces in which people see total eclipses more rebel more often?

It turns out that they do.

And it’s not a tiny effect. The researchers pin down how much more likely this is, writing that “counties in the totality zone of a solar eclipse are about 18 percent more likely to experience a rebellion in the eclipse year relative to counties that are outside of the totality zone.”

Just to put that into modern context, one might wonder which has a bigger effect on rebellions, eclipses or the massive social media platform in modern China, Weibo. Qin et al. looked at exactly this question. They find an effect similar to the one that eclipses produce: “We estimate that, because of Weibo, the occurrence of a protest in one city increases the probability of a protest taking place within two days in any of the other cities by 17%.”

So, eclipses are a tiny bit more powerful than social media.

How can this be?

Now, everyone knows that you can’t logically infer causality from a correlation—in this case, between rebellions and the frequency of total eclipses.

Still, it’s not that easy to come up with a third variable that might cause both. What terrestrial force could move the moon? The authors do run through some other possibilities, including the idea that those regions differ in some way due to their climate or history, but the finding is robust to these possible confounds.

So what’s going on?

If a bunch of people want to rebel, there are two different but related problems: a public goods problem and a coordination problem.

The public goods problem rests on the decision each potential rebel faces. There is danger in rebelling, and each person would like to enjoy the benefits of a successful rebellion without putting themselves at risk. This problem is difficult to overcome, especially when the cost of participating is the possibility of death. So everyone wants others to rebel without having to take part.

The public goods problem is easier to solve as the number of participants in the rebellion increases. As the crowd increases, each person is less likely to be hurt. Greater numbers increase the chance of winning, and larger forces are likely to sustain fewer casualties in combat. So the trick is to get everyone out and protesting in the same place at the same time. They need to coordinate, not unlike the cicadas I’ve discussed in the past.

Which brings us back to Thomas Schelling.

One of my favorite lines from Les Miserables, the musical, is in the song “Red and Black.” It’s about coordination:

We need a sign
To rally the people
To call them to arms
To bring them in line!

Schelling points are all about these signs. This plugs into the argument in Steve Pinker’s new book, When Everyone Knows That Everyone Knows. There are three rungs to the knowledge ladder.

1. I know a thing.

2. I know that you know the thing.

3. I know that you know that I and everyone else knows the thing.

The importance of the second and third levels explains the usefulness of very visible events. Very visible events don’t just convey information; they generate common knowledge, or beliefs that everyone knows others also have. When an eclipse occurs, I don’t just learn that something strange has happened in the heavens—I learn that millions of others have observed the same thing at the same time, and that they know that I have observed it too. As beliefs climb this ladder, coordination becomes dramatically easier.

Eclipses are really nice Schelling points—coordination points—because they are high up in the sky. The fact that they are rare is an asset, not a liability. If they happened all the time, then people wouldn’t be able to coordinate on any given eclipse.

So how do you use an eclipse to foment a rebellion? One idea is to spread the belief that an eclipse means that change is coming, or a leader is about to fall ill, or really any sort of belief that makes the eclipse seem like a good time for everyone to rebel. It is not surprising, then, that there is evidence of these sorts of beliefs in China during this time. And so, during total eclipses, people believed that others saw the same thing—since it was high in the sky—and this created the focal point for a rebellion to begin. A sign, to rally the people, to bring them in line!

Rulers knew these ideas were out there, and there is evidence that they took precautions. For example, another recent paper, by Chen et al., suggests that during Chinese history, “emperors adopt an eclipse-dependent tax-cut strategy to reduce farmer rebellions.”1

The role that Schelling points play in coordinating action by many explains why places with more dramatic eclipses—locations where the eclipse is total rather than partial—rebelled more often. In those places, potential rebels knew that others saw the same dramatic dance in the sky, giving them greater confidence that others would take to the streets as well. In places with fewer eclipses, or places in which eclipses were only partial—the extent of an eclipse depends on where you are—people had less confidence that others would see the event as sufficiently dramatic to motivate rebellion.

Today, social media plays the same sort of role. When Elon Musk tweets something out on X, it’s a good bet that A) many people read the tweet and B) people who read the tweet believe that other people have read the tweet. Viral posts have the same property. If you see something on social media with millions of views, then you have evidence that many people have seen it.

The fact that social media platforms create common knowledge explains why they foment rebellions in the same way that eclipses do. They are both solving the same problem, the problem of common knowledge, or coordination. In the past, you could stand in the town square and yell, hey, let’s kill the baron and build a government of and by the people or whatever, but your reach was limited to how far your voice carried. Now, of course, your voice can reach millions, as tweets and other posts are amplified.

It’s important to note that the issue isn’t just that social media gets information out there so that everyone has access to it. There must be common knowledge. The usual example of this is The Emperor’s New Clothes. Everyone sees that the king is naked. But no one dares say anything until the child takes the belief to level 3, common knowledge. Broad dissemination of information is not in itself sufficient.

The paper I linked to above, “Social Media and Collective Action in China,” shows how social media posts spread protests and strikes. The fact that the research targeted China is important. There, the government, well aware of the power of the internet in general and social media in particular, exercises control over content, censoring material that might be dangerous. Nonetheless, even with these controls in place, these platforms do manage to foment unrest.

Governments have been aware of the role that social media might play in rebellions right from the start. For an excellent account, see Sarah Wynn-Williams’ book, Careless People: A Cautionary Tale of Power, Greed, and Lost Idealism. From her front row seat at facebook, she witnessed how governments began to see the threat posed (and opportunity created) by facebook and other social media platforms and reacted to it by working to control their power to bend them to their own ends. This trend continues to this day, as authoritarian regimes suppress the spread of information during times of instability by restricting the use of some or all of the internet for its citizens.

Now, what about the strong claim in the title, that eclipses caused rebellion. Causality is a notoriously difficult concept to establish. And, of course, the fact that rebellion correlates with eclipses doesn’t, in itself, establish causality.

Still, the fact that some regions experienced total eclipses more than others is pretty close to random assignment. Those regions might be different from other regions in systematic ways, but eclipses don’t respect human-drawn boundaries, so that doesn’t seem too likely. And, of course, it’s not just that those regions rebelled more often, it’s that they did so around the time of eclipses. This finding suggests that if it hadn’t been for the eclipse, there would have been no rebellion, showing but-for causality. But for the eclipse, the rebellion would not have occurred.

Rebellions are difficult without a Schelling point. Recently, the Wall Street Journal published an editorial by Babak Seradjeh about the massive demonstrations in Iran. Referring to the son of the former ruler of Iran, Shah Mohammed Reza Pahlavi, Seradjeh writes: “Mr. Pahlavi called on Iranians to turn out, and they responded en masse. My family in Tehran described a stream of people suddenly pouring into the streets on Jan. 8 at 8 p.m., the exact time of Mr. Pahlavi’s call. The arteries of the capital and other major cities pulsated with people. The regime cut off the internet and responded with unprecedented violence.”

In this brief passage, we see all the pieces to the puzzle: the attempt to provide a Schelling point for rebellion, people answering that call, presumably believing that others will as well, and the crackdown by authorities to undermine coordination through social media.

Chen et al. suggest that the relationship between celestial events and rebellion has deep historical roots, writing that in “the Bamboo Annals (zhushu jinian), a chronicle of early China, recorded an eclipse in 1059 BC that was interpreted by a vassal king, Wen of the Zhou Dynasty, as a sign to challenge his cruel overlord.”

Humans have been looking to the sky for Schelling points since time immemorial.

Today we’re trying a new format called Bone Shards, named after the partial fragments often found at fossil dig sites. These posts bundle together two to four quick-hitting pieces that aren’t long enough to justify a full post, but are too substantial for a Note.

Please let us know if you have strong feelings, for or against, this kind of post. As always, thanks for reading and engaging.

The Limits of the Satiation Model—JZ

Rob likes to tell a story as part of his gripe about explanations:

Many years ago, I was having dinner with a collection of Very Smart People, the sort that get invited to scientific gatherings where academics try to impress one another and eat free meals.

It was at one of these free meals—I think it was a schnitzel place?— that I asked the gathered worthies if people eat faster when they are hungrier. I myself certainly do this, but I was loath to generalize from my very small sample size of little ol’ me.

The table agreed that this was, indeed, the case, and I wondered aloud why. The senior Very Smart Person, to whom everyone’s gaze had turned, squinted his eyes slightly and, after a dramatic pause, said, “drive reduction.” There were pious murmurs of agreement, and some considered nodding at this and the discussion passed on to the important matter of whether the conference organizers would object if we ordered another round of Weissbier for alles.

Rob tells this story to show that a word—or even a pair of words—can’t constitute an explanation. One must be more specific than that. A not-specific-enough explanation becomes, as Gerd Gigerenzer writes, “a Rorschach inkblot,” into which anyone can read what they wish.

I’ve moved in the direction of agreeing with Rob, but still maintain that “drive reduction” is a perfectly reasonable answer in that context. It seems clear enough what the senior Very Smart Person meant: as someone eats, thereby reducing the empty space in their stomach, the drive based on that empty space (hunger) also diminishes. Taking it a step further, it seems plausible that as hunger is experienced at lower intensity, one eats more slowly, since the problem has become less urgent. Similarly, the more intense an itch, the faster I would expect someone to bring their fingers to the spot.

This satiation model, in which motivation declines as a need is progressively satisfied, describes plenty of phenomena, from eating to taxi-driving behavior. Work in economics shows that taxi drivers having a great night with lots of fares tend to quit earlier than those having a bad night. Those having a good night “reach their fill,” so to speak, while those having a bad night stay out longer, still “hungry” for fares.

Then there are other phenomena that don’t fit the satiation model at all, and yet the model is commonly applied anyway. For example, what do we do with the fact that:

• I experience less psychological resistance at the prospect of a second workout in a day than the first

• Those who have had more sexual partners before marriage have a higher likelihood of cheating

• Those with abundant power, money, and championships seem to want more of those things than those with less

• I feel a stronger urge to write—and have more ideas to write about—than when Living Fossils launched nearly three years ago1

These don’t seem explicable by the satiation model, do they? Yet we often invoke that model when thinking about them. According to satiation logic, one might reason:

• You’ve already hit your quota of workouts for the day; relax

• You’ve sown your wild oats; time to settle down

• You have more power/money/championships than you know what to do with. Why do you want more?

• Surely, with sixty articles behind you, you’re running out of things to say…

I tried explaining the inadequacy of the satiation model in a prior note, arguing that a better explanation for these patterns is plain old positive feedback. You work out in the morning, feel great, and later think: let me hit that serotonin again. You publish an article, someone leaves a kind comment, and you’re even more motivated to write the next one.

But I don’t think I went far enough in that note because I didn’t explain why the satiation model is so tempting despite not being a great explanation. It’s tempting because it’s partly right.

More precisely, the satiation model describes short-term stop signals, but it fails as a theory of long-term motivation because motivation over time is expectation-based rather than consumption-based.2 If long-term desire were consumption-based, the richest people would be the least motivated, the most successful writers would run out of things to say, and Michael Jordan’s first retirement might have stuck.

Instead, when something happens with regularity, we begin to expect it. We quietly build it into our model of the future. And because satisfaction depends, at least in part, on how outcomes compare to expectations, yesterday’s pleasure quickly becomes today’s baseline. To feel pleased again, the outcome must exceed what we now take for granted—what we now expect.

Let’s return to taxi drivers. The satiation model does a good job explaining their behavior on a given night: those who hit their number quit, while those who haven’t keep driving. But a different set of mechanisms is needed to explain their behavior over time. A driver who reliably hits their number might stay out even longer on nights when they fall short—and, more importantly, may begin to raise the number itself. Indeed, where does that number come from, if not (at least in part) from prior experience?

The mechanisms needed to explain long-term desire include: positive feedback, habituation, tolerance, diminishing returns, prospect theory, and hedonic adaptation. Together, these explain why more—or better—becomes the price of feeling the same.

The satiation model helps explain why people stop in the moment, but not why motivation changes over time. It cannot explain why those with more often want more, or why those with less might come to want less. For that, we need the insight that satisfaction does not extinguish desire so much as inflate it: as people grow accustomed to being satisfied, the threshold for satisfaction rises. All else equal, Michael Jordan—a perennial champion—may have wanted his next championship more than Charles Barkley, who never won one.

Why do you get what you pay for?—JZ

The familiar saying “you get what you pay for” is usually taken to mean that higher prices buy better products, services, or experiences. Most people explain this in straightforward economic terms: quality costs money. Better materials cost more, better workers command higher wages, and so on.

All this is true, but it ignores a second way money dictates value: social expectation.

I stumbled on this second way just this morning, after yet another conversation with the longwinded, cantankerous, and supremely well-meaning Mrs. T., one of the staff at my daughter’s daycare. Now, I don’t have a bona fide complaint against Mrs. T.—not yet, anyway—but if I ever did, I’d be constrained in voicing it by the fact that my wife and I pay downmarket rates. A request that Mrs. T. clean up her act would fall into the same category as a request for organic baby food or for the place to smell less like diapers. Hey, guy—what do you think this is? The Ritz?

I’d be constrained, that is, by the very notion that one gets what one pays for. I can only be a tyrant if I’m paying tyrant prices.

Here’s another way to think about it. For some things, like organic fruit, the price more or less is what it is. There’s limited room for illusion. But something like worker quality is different. While there are real differences among workers—stemming from personality, training, or experience—the larger determinant of how people behave is often social norms, with price playing a central role.3 Put a low-end daycare worker into a high-end environment, and you’ll often see rapid improvement as she steps up to meet the prevailing expectations. This is where ordinary economics gives way to social economics.4

Don’t get me wrong: wages matter in attracting higher-quality workers. But they also do a great deal of work enforcing social norms. Were I to pay double for my daughter’s daycare, I wouldn’t only be paying for better-trained staff; I’d also be paying for the right to demand higher standards. Basically, I’d feel more comfortable suggesting to Mrs. T. that she should shut her beak once in a while.

Daycare is not the only setting where this applies, nor is price the only norm-setter. Another example comes from sports: research shows that strong teams do indeed play down to weaker opponents. The weaker team sets a standard of engagement that the stronger team cannot help but respond to.

Long story short, until Rob and I turn on paid subscriptions, I’ll have no choice but to suffer Mrs. T in silence.

The Wisdom of My Grandfather—RK

I have a complicated relationship with wisdom. It comes in three pieces.

First, my maternal grandfather – with whom my relationship was not especially complicated – used to say, “People grow too soon old and too late smart.”

I’m sad to say that I don’t recall many of my grandfather’s sayings, so the fact that that one stuck with me seems significant.

Second, there’s Shani. He has cornered the market on wisdom, and I have come to appreciate, well, his wisdom. He says wisdom is Important. He’s usually right.

Third, there’s this evolutionary piece. Lots of evolved creatures come with lots of True information equipped standard. You pop out of your chrysalis, and boom, you know how to fly.

If wisdom is important, and guides good decision making, then why do we wind up getting too soon old and too late wise? Why doesn’t wisdom come with our standard equipment? Why didn’t evolution craft human minds to be smart, or at least wise, from the get-go?

This idea is embodied in stereotypes of the Old Wise One, a staple of fiction, especially fantasy. It also accords with my experience. With the wisdom that comes from being in my mid fifties, the youngs often appear to act unwisely. As, no doubt, I did.

I don’t really have a good answer for this. My guess is that there are a few key pieces.

The first is that wisdom isn’t quite as wise as we think it is. That is, suppose you measure once instead of twice. Ok, yeah, much of the time it’s probably a good idea to measure twice so that you reduce the chance of error. But not always. There are occasions when time is so pressing, it’s actually better to measure once and hope for the best, even if that means cutting twice. So wise sayings aren’t universally applicable.

Second, and related, the value of some aspects of wisdom might change with age. It seems wise to recognize that “if you chase two rabbits, you catch none.” That is, focus of attention on one thing improves your chances of accomplishing your goal. But, I dunno. When you’re younger, you can focus on more than one goal at a time, often without much loss to the pursuit of either. My experience as I get older is that I have to concentrate on one thing to get anything done.

Third, to take a combination of one of my favorite things—it’s complicated—and one of Josh’s favorite things—mismatch—it could be that the modern world is too complex for ancient wisdom to work. So maybe a stitch in time saves nine, usually, but tradeoffs depend on a lot of factors. Maybe sometimes it’s not worth putting in the time to do repairs in the face of other exigencies.

In short, maybe—probably, I’d say—the nuggets of wisdom that turn out to be the most valuable vary from time to time and place to place. So evolution couldn’t just build in principles of wisdom because what principle works in a given situation varies too much from context to context. What is wise depends on the situation you’re in, and situations have varied tremendously over evolutionary time. Worse, the wise move might change during your lifetime as culture changes and you yourself change.

Are you in a situation in which you should look before you leap or should you strike while the iron is hot? It depends.

Maybe, as the Serenity Prayer intimates, wisdom is about learning how to tell the difference. Humans are pretty good at learning, so eventually, yeah, we learn to tell the difference. But only after we’ve put in the time. Something that comes with age.

Not long after finishing the series, I had a paradigm-shifting therapy session. I try not to throw that term around lightly.

The client, Mark, had initially reached out for help with existential anxiety. In fact, he contacted me specifically because he’d read The Existential Relief of Having Children and felt it spoke directly to his situation. On the verge of becoming a father, Mark had begun to question everything. He felt uncertain about his career and unmoored from any clear sense of purpose. He suddenly regretted how he had spent his life, even while acknowledging that it had been fun and meaningful. He didn’t know who he was anymore—or rather, he worried that he could not continue being the person he had been.

For the first few months of our work, that framing served as our guide. We assumed Mark was in a kind of midlife crisis, brought on by an impending life change and a lack of prior self-reflection. Mark felt he had stumbled into a career, a marriage, and an identity without ever stopping to assume The Thinker’s pose. Now the big questions had arrived all at once, and it seemed our task was to take them on one by one.

Curiously, although I liked Mark, I approached our sessions reluctantly. The task of rebuilding him from the ground up felt heavy and boring. We talked at length about his past, present, and imagined future. He wrote lists of values and goals. He kept a thought journal. We mulled who else he might talk to—perhaps his wife—to gain further clarity.

In short, we did all the things one does when one assumes that the problem is internal and that the solution lies in insight. And we did it urgently, under the shadow of Mark’s suffering and his wife’s looming due date. I think this is why, although I usually enjoy the rambling nature of therapeutic conversation, I frequently wanted to rein Mark in. I was hyper-aware of how much work we had to do and how little time we had to do it.

But something felt off. We didn’t seem to be making any progress.

Eventually, it occurred to me that situationism might offer a better lens. (Being knee-deep in the series certainly helped.) Perhaps the problem had less to do with Mark than with his circumstances.

Crabs in the Canopy

In my musings on mental health, I frequently return to images of animals radically out of place: crabs in the canopy, otters in the desert, armadillos in the arctic. What would we learn by studying such animals? One obvious lesson is that if we wanted to help them, we should send them home. It would be a waste of time—or worse—to focus on what they were thinking, feeling, or doing, since these would be the downstream outputs of an animal in the wrong environment.

For example, we could draw all sorts of conclusions by watching a crab scrabble at the smooth surface of a tree branch. Perhaps there is food just under the bark. Maybe it’s a mating ritual or territorial instinct. The correct explanation, of course, would be that the crab is trying, unsuccessfully, to burrow in the sand. Yet we’d have a hell of a time figuring this out if we didn’t already know that crabs belong on the beach.

For a nonliving example, consider the vacuum cleaner. No one would claim a vacuum is defective because it fails to wash dishes, nor would a manufacturer reimburse you because it performed poorly while clearing gutters. Only when a product is used for its intended purpose does it make sense to say that it isn’t working properly. Otherwise, the problem is not the tool but the way it is being used.1

Humans, of course, are crabs in the canopy. We are vacuum cleaners on the roof. This modern world we have built, in the blink of an evolutionary eye, is not our ancestral home. So we live in many ways for which we were not designed—for example, in the constant presence of so many strangers. Yet few people carry this perspective with them, even into fields like therapy, where you might expect it to be foundational. The result is that when distress appears in its various confusing forms, therapists and laypeople rarely treat it as situational pathology—as the noxious byproduct of a poor fit between person and environment.

Instead, distress is usually taken at face value. A client says they are feeling worthless and the therapist gets to work investigating what about the person or their past might have given rise to this feeling. Eventually, the therapist will try to convince the client that they are not worthless, even if it means revealing their own high regard for the client. But the therapist might be missing the forest for the trees here. The main value of the client’s feeling might lie in the more general signal that something is off—but what, exactly, requires further investigation.

The point is that when the environment isn’t right, subjective content is low-quality data. After all, if we found a way to ask the crab how it felt scrabbling tree branches, it would presumably find various ways to say “bad.” Probably none of these formulations would lead us to the conclusion that we ought to transport it to the beach unless we knew that separately—or made the rare intellectual move of zooming out and contemplating a bigger picture.2

Let’s consider one more clinical example before returning to Mark. Suppose a client begins reporting weekly nightmares, many featuring a distorted teddy bear from childhood. This client and I could focus on what the teddy bear symbolizes, or why it appears in altered form. Or we could simply step back and notice that her job is putting her through the ringer. The specific content of the dream may be vivid and emotionally charged, tempting both her and me to interpret it, but in many cases those details are irrelevant. The more salient fact is that the client is having stressful dreams, most likely because of a stressful work environment.

I’m not saying that subjective experience is useless or meaningless; I’m asking about the best way to understand and use it. For far too long, therapy has been the practice of overvaluing subjective content and undervaluing ecological or situational fit.

Mark and My Turning Point

It was against this backdrop that Mark and I had our breakthrough session. I try not to throw that term around lightly.

We concluded—tentatively at first—that focusing on the content of his thoughts was not merely unhelpful, but could be actively counterproductive if taken too far. Over-attending to them kept him mired in misery for three reasons. First, there was the opportunity cost: time spent ruminating was time not spent engaging with life in ways that might help. Second, attention is not neutral. To focus on misery is, in a real sense, to be miserable. Third, existential questions are precisely the sort the human animal is poorly equipped to solve; they invite rumination without resolution.

We began thinking of Mark’s negative thoughts and emotions—the “toxic cloud” that followed him on bad days—as meaningful primarily as an indication that something was wrong in the ecology of his life. At the same time, we remained open to the possibility that the specific content of his thoughts, feelings, and dreams might tell us exactly what that problem was.

With this shift in perspective, Mark and I were able to diagnose the issue in a single session. So what was happening?

Mark commuted to work Tuesday through Thursday and spent Friday through Monday at home. The toxic cloud tended to form on Fridays, peak on Sundays, and evaporate by Wednesday. It seemed to appear after a kind of “permission timer” had expired. Mark was productive Tuesday through Thursday, thereby “earning” leisure, but by Sunday that credit had disappeared. He began consuming relaxation he did not “deserve.” He felt increasingly worthless for being at home, and even activities he had long enjoyed—like golf—brought him no joy.

“So it’s almost as if being home doesn’t distract you enough from the toxic cloud,” I said.

“Exactly,” he replied, before pausing. “Only I don’t like the word distraction because it sounds like I’m avoiding my life.”

“How about engagement?” I suggested. “When you’re engaged with life outside the home, you don’t have the bandwidth to ask these questions. That’s the thing about home: it’s comforting and relaxing, but it also allows you to ruminate.”

Mark thought that fit much better.

Our hypothesis was all but confirmed when I asked him about the COVID lockdowns. During that period, he had been home continuously, so the toxic cloud should have been omnipresent, right? Quite the opposite. Mark said he cherished that time, and as someone who had cherished it, too, I knew why. There was no expectation that anyone should be doing anything. As Rob would say, the opportunity cost of doing nothing had dropped to nearly zero.

This contrast helps explain why the cloud appeared when it did. In effect, the opposite of lockdown had occurred: Mark learned he was going to be a father. A father, unlike someone trying to avoid spreading a disease, is expected to be doing something—often a great deal. The role carries heavy social expectations. A father is supposed to be productive, self-sacrificing, mature. He is not supposed to loaf or disappear into leisure.

As a father-to-be, Mark’s opportunity cost of doing nothing rose sharply, provoking a cascade of questions he couldn’t answer—and, fortunately, didn’t need to.

The Solution

So what did Mark need?

First, he needed his son to arrive and tell him what to do. Parenthood would impose the structure he clearly craved. It would deliver that clarifying necessity.

In the meantime, he needed to force himself back into the world. Being at home—even to work—was not lighting up his brain’s productivity sensors. Much as he didn’t want to, he needed to earn the leisure he was taking, and that meant being out and about.

This was difficult for Mark to accept. He is someone who needs leisure and solitude, and he rightly recognizes four uninterrupted days at home as universally coveted. But too much of a good thing is still a bad thing—and, more importantly, human wants are not always trustworthy guides in a modern world that increasingly takes advantage of them. For Mark, working a full week in the office, and thus earning the weekend, had long sufficed. I encouraged him to return to something close to that arrangement.

“Compare the feeling of burnout or social fatigue you get from being among people, commuting, running errands, and so on,” I said, “with the toxic cloud we’ve been talking about.”

“Oh,” he said. “No comparison. The toxic cloud is a hundred times worse.”

“Then err on the side of social fatigue.”

My hope is that once Mark gets the ratio right, the toxic cloud will largely dissipate—and with it, the thoughts that had once seemed so important to both of us.

Conclusion

Consider how many standard therapeutic interventions would have been wasted—or actively harmful—in this case: journaling, goal-listing, open-ended exploration of feelings, deep dives into childhood and family history, bringing other people into the dark thoughts…

Really, all Mark needed was to be led out of the canopy and onto the beach. I’m just glad I realized this sooner rather than later.

Keep in mind: who Mark is still matters. A person–situation fit always depends partly on the person. It’s just not usually the person we rely on to change. The situation is the operative variable because it is typically easier to change than one’s personality, and more reliable than depending on personal capacities like insight or willpower to do the heavy lifting.

Another brief example from my practice will help make this concrete. A client, Dewey, was struggling to stay productive. The computer—YouTube in particular—was consuming much of his time. I asked whether he had found anything that helped him focus. He said yes: the eyes of other people. This stuck with me because I’m the opposite. In fact, my wife occasionally mourns that we don’t read together, but the truth is that any human presence requires me to devote a portion of my attention that I would rather devote to what I’m reading or writing. For me, work requires solitude. Even a window to the street, allowing people to peer in, is too much.

So yes, individual differences matter. But they are not the operative variable. It is far easier and more realistic for me to work alone, or for Dewey to work under the motivating presence of others, than for either of us to reengineer our personalities. I’m sure we could adapt if we had to—but we don’t.

Mark’s misery was not rooted in childhood, personality, or unexamined values. It arose from having too much “unearned” time when he was socially expected to be getting serious—to be growing up. If our conceptualization is right, Mark’s discomfort will be resolved not by answering his questions, but by earning the leisure he takes through more fully engaging with the life around him. This conclusion ran against both Mark’s instincts and my training.

Again, when the situation is wrong, subjective experience is low-quality data. Sometimes the most humane thing a therapist can do is stop listening so closely and just get the crab to the beach.