Today we are looking at some bad mental models and the harm they cause. Let’s begin with what must surely be a contender for the anti-award of most harmful one. Responsible for zero-sum economic thinking, and all the bad policy that can flow from that. I take my inspiration from Paul Graham’s essay How to Make Wealth, where he says the mistake stems from a confusion between wealth and money, and speaks of the Pie Fallacy. Let me quote one particular bit:
When wealth is talked about in this context, it is often described as a pie. “You can’t make the pie larger,” say politicians. When you’re talking about the amount of money in one family’s bank account, or the amount available to a government from one year’s tax revenue, this is true. If one person gets more, someone else has to get less.
I can remember believing, as a child, that if a few rich people had all the money, it left less for everyone else. Many people seem to continue to believe something like this well into adulthood.
What other contenders can we think of? What other common mistakes? Robin Hanson might note our frequent lack of awareness of hidden motives, to do with prestige and status. How much harm does that unawareness do? Or could it be even beneficial instead, maintaining “useful fiction”? I would probably guess it’s harmful, conceivably very much so. But then, taking something too naively at full face value, that’s not really what I had in mind when setting out to blame bad mental models such as the pie.
Gary Taubes, for his part, might note the “calories in, calories out” (CICO) fallacy. And there is a lot at stake, with obesity, or at least the so-called “Western diseases” associated with it, like diabetes, considerably shortening our lifespans. Even though that effect is masked by successes against infectious diseases and progress in health technology. Let us look at the fallacy more closely, then. It’s been a few years since I read Taubes’s excellent book Why We Get Fat, which all the CICO-related material here ultimately draws from; so it is perhaps worth stating explicitly in this case that all errors and distortions to follow are (of course) my own.
The First Law of Thermodynamics, or energy conservation principle, cherished by CICO adherents is not really such a “tangible” mental model as the pie. At least not if left uninterpreted. In which case it would be a strange recipient for the anti-award anyway as, obviously, it is not wrong! The mistake creeps in only when we impose a certain causal interpretation. Yet why do we do that? When children grow, we don’t think that this is because they consume more calories than they expend. Rather, we somehow think of them as being in “growth mode”, and that’s why they get hungry and consume many calories. So shouldn’t we equally have the wherewithal to think of people as being in “excess-fat-accumulation mode”, as opposed to “eating too much” and “exercising too little”? No fundamental overturning of intuitions is needed, of the kind that would be for, say, quantum physics. Hence I submit that the standard understanding of CICO causality is fixed in place actually by a bad mental model: the body as a sack, that bulges when you fill it.
Why is a reverse understanding of the causality preferable? For instance, Taubes points to rat and mouse models in the study of obesity, where these rodents get fat even if their food intake is restricted. The same holds for food-restricted ground squirrels in late summer, when their bodies prepare for hibernation and will virtually always find ways to stash away the calories. He also points to overweight mothers with visibly underfed children. Or shall we believe that the women consumed excess calories while their children were starving? Again, he points to progressive lipodystrophy, a rare condition where part of the body is subject to fat accumulation and another part actually loses fat. If an accumulation area covers x percent of the body, most people would not blame calories as long as x < 100. But then immediately do so once x = 100. How can that make sense?
In people’s defence, those are empirical points based on data not widely known. Many of us never heard of lipodystrophy. Compare that to the pie fallacy, which is so easy to refute. As Graham notes in his Wealth essay, you could do nothing next summer, or you could restore your beat-up old car. You could even go around smashing windows, one might add. There is no conservation principle here. No fixed pie. Wealth can be created or destroyed.
Then again, some refutations of the sack fallacy almost work from first principles also. Consider the idea of storage as commonly conceived in this context, in line with the sack model and used to explain why our bodies wouldn’t simply shed a surplus. If we eat a lot, they will use that windfall to store calories for the lean times that were common in our evolutionary past . . . but relying on leaner times to guard against getting fat and diabetic, isn’t that a bit risky? Wouldn’t we occasionally have encountered prolonged spells of plenty in the past? And if evolution designed me to grab storage opportunities as they arise, how come I am sometimes not hungry even if there is food left in the fridge, or for that matter if the supermarket is still stocked and I am privileged enough to be able to afford the food? Meanwhile, as for the exercise side of the CICO coin, if the fat man follows recommendations and plays badminton to burn calories, he loses weight, supposedly . . . but what about his slim playing partner? Will she become underweight? No, she will be tired or hungry afterwards and end up keeping her weight — but why then should the same not apply to him?
Little of it makes sense. And I didn’t even mention here all of Taubes’s ways to pierce what I called the sack fallacy. The sack deserves a nomination, against the pie, for the anti-award of most harmful mental model. Now, it starts with one handicap, though. Whereas the pie just needs to deceive ordinary people, who are then apt to demand policies that sabotage wealth creation, the sack needs to deceive the experts, who are then apt to fight obesity in the population with behavioural advice of an unproductive kind. Where instead they should be investigating what poisons us and causes the excess fat accumulation. So could the experts, as opposed to naive laypeople, have been deceived?
Indeed they could, long and comprehensively enough for a vast epidemic of obesity and diabetes, and other Western diseases, to unfold. Taubes shows that. For illustration, just check the quotations from American, British, French and German health authorities in the footnote on page 6 of Why We Get Fat. (That’s from 2010, but for an impression of how the sack fallacy is still keeping American authorities in its grip, look here.)
The sack turns out to be a viable competitor to the pie. What ought to follow now is some attempt at quantification of the harm done by each. But would I possess the expertise to estimate that? I’d prefer to find and apply some crude tie-breaker, for this laziest of ties, and leave it at that. And perhaps one might be found, we will see, if we look at the two contenders specifically in a Western-versus-global context.
To start with, such a context shows particularly well a caveat. It’s not that the insight “wealth can be newly created” would then suddenly imply “all wealth is newly created”. When in the age of colonialism Western countries stole wealth from others, the pie model had that covered. On the other hand, the danger is clear as well. What could be more natural than assuming that all the wealth disparities are due to colonial robbery, perhaps even ongoing, if one’s thinking is beset by the pie fallacy? One will never get around then to consider certain assets of today’s Western countries that are highly relevant here, like relatively decent governance, social capital, rule of law, etc.
Indeed, going by Graham’s Wealth essay, it is such features that really render the pie model a consequential fallacy. It was only in mediaeval Europe, he writes, that opportunities began to arise to create wealth that wouldn’t then be quickly appropriated by those in power. Perhaps in corrupt, badly-governed countries the pie model of wealth is adequate enough? Moreover, if the relevance of the pie fallacy is somewhat specific to the Western world, the same can be said about the sack fallacy — the term is “Western diseases” after all.
However, those diseases are exported around the world! Whereas Western practices of good governance, alas, don’t seem to spread so well. Along this line, one could say that the sack fallacy is going global, whereas the pie fallacy is staying relatively more confined to certain rich countries. If I just gloss over the ways in which this assessment is questionable, could it serve as the excuse I was looking for to give the SBP anti-award to the sack?
For let’s admit it, I was always biased towards it over the pie. I hope that’s not how it works with real anti-awards. Perhaps I have particular respect for the sack because I was myself fooled by it so easily for so long. But in the end, despite my bias, the result has to be a different one. Why? What happened? Very likely, I will have missed important examples of harmful mental models here, but at least I eventually managed to acknowledge a third contender, besides sack and pie: the slate, as in The Blank Slate: The Modern Denial of Human Nature, a 2002 book by Steven Pinker which, according to its Wikipedia page, contains some serious accusations. Five are listed, at the time of this writing:
Totalitarian social engineering. If the human mind is a blank slate completely formed by the environment, then ruthlessly and totally controlling every aspect of the environment will create perfect minds.
Inappropriate or excessive blame of parents since if their children do not turn out well this is assumed to be entirely environmentally caused and especially due to the behavior of the parents.
Release of dangerous psychopaths who quickly commit new crimes.
Construction of massive and dreary tenement complexes since housing and environmental preferences are assumed to be culturally caused and superficial.
Persecution and even mass murder of the successful who are assumed to have gained unfairly. This includes not only individuals but entire successful groups [...]
Hard to see how the slate could not top both sack and pie, in fact! However, I notice that two of these items, psychopaths’ release and dreary tenement complexes, do not exactly sit at the “populist” end of the spectrum. Even in the case of Marxist views of human nature I have my doubts as to how pervasive they are outside academia and the world of intellectuals. To keep the slate down, I modify the assessment criteria: we will now take into account, not just harm caused, but also pervasiveness (in the general population). In my defence, what I wanted to get at was something like innocent and addressable common mistakes, as opposed to concepts which are needed (or at least lazily invoked) by ideological movements but perhaps not really that much of a hazard to most “ordinary” people. Try to ignore for a moment all the bad taste here (an “award” bestowed for severe harm): there is also a more constructive point, namely that the most harmful mistakes should be most urgently addressed, in order to make the world a better place.
Still, against my attempts to prevent the slate from taking over, the fifth of the items above looks uniquely dreadful. Let me spell out which persecuted successful groups Wikipedia goes on to note. “Examples include Jews in Nazi Germany during the Holocaust; kulaks in the Soviet Union; teachers and ‘rich’ peasants in the Cultural Revolution; city dwellers and intellectuals under the Khmer Rouge.” Could anything be possibly worse than the slate?
Then again, why would Pinker really blame it here? When Wikipedia says, as quoted above, “the successful [...] are assumed to have gained unfairly”, this would seem to come straight from the pie’s playbook instead! Admittedly, I haven’t read his book, but let me quote from a preview Pinker published in Discover magazine, October 2002. For one thing, the blank slate “became the official doctrine among thinking people only in the first half of the 20th century”. It did so not for bad reasons at the time; but what I suspected about ordinary people versus intellectuals appears confirmed. Then, towards the end of this text, pace Wikipedia, Pinker does not connect Nazism to the slate (on the contrary), just the others. With emphasis as in the original:
Nazism was not the only ideologically inspired holocaust of the 20th century. Many atrocities were committed by Marxist regimes in the name of egalitarianism, targeting people whose success was taken as evidence of their avarice. [...] Although both Nazi and Marxist ideologies led to industrial-scale killing, their biological and psychological theories were opposites. [...] Marx did not explicitly embrace the blank slate, but he was adamant that human nature has no enduring properties: “All history is nothing but a continuous transformation of human nature,” he wrote. Many of his followers did embrace it. “It is on a blank page that the most beautiful poems are written,” said Mao. “Only the newborn baby is spotless,” ran a Khmer Rouge slogan. This philosophy led to persecution of the successful and of those who produced more crops on their private family plots than on communal farms.
Yet I still don’t quite understand the path from slate to atrocities in these cases. If anything, should the slate not divert anger into friendlier treatment — recall the psychopaths above? The pie, by contrast, indeed looks implicated. But if such mass murders are associated with it to a considerable extent, and plausibly they are, then I must conclude it is the pie after all that is the most harmful mental model in human history.