Just another run-of-the-mill, middle-of-the-road Pinker volume. Which is another way of saying it’s bloody marvelous. What a consummate intellectual this man is! Every one of his books is a bracing river of fluent readability to delight the non-specialist. Yet each one simultaneously earns its place as a major professional contribution to its own field. Grasp the fact that the field is different for each book, and you have the measure of this scholar. Steven Pinker’s professional expertise encompasses linguistics, psychology, history, philosophy, evolutionary theory—the list goes on. There’s even a book on how to write good English—for, sure enough, he is a master of that too.

And now Rationality. In the final chapter of this book, Pinker raises the serious, even baffling, question of how it could ever have been necessary to mount an argument against slavery, against drawing and quartering, against breaking on the wheel, or against burning at the stake. Isn’t the case against slavery obvious? Nevertheless, it had to be made. One could ask the same question about the justification of rationality. How is it possible that rationality needs an advocate? Isn’t it obvious? Who will mount an argument against reason? Peter Medawar, elegant patrician among Nobel scientists (he should have won the Literature prize as well) came close, but he was being satirical: “The official Romantic view is that Reason and the Imagination are antithetical, or at best that they provide alternative pathways leading to the truth, the path of Reason being long and winding and stopping short of the summit, so that while Reason is breathing heavily there is Imagination capering lightly up the hill.”

Pinker puts the point in the argot of today: “Rationality is uncool. To describe someone with a slang word for the cerebral, like nerd, wonk, geek, or brainiac, is to imply they are terminally challenged in hipness.”

And he continues:

For decades, Hollywood screenplays and rock song lyrics have equated joy and freedom with an escape from reason. “A man needs a little madness or else he never dares cut the rope and be free,” said Zorba the Greek. “Stop making sense,” advised Talking Heads; “Let’s go crazy,” adjured the Artist Formerly Known as Prince. Influential academic movements like postmodernism and critical theory (not to be confused with critical thinking) hold that reason, truth, and objectivity are social constructions that justify the privilege of dominant groups.

Only the final member of this list is actively pernicious. The influence of “postmodernism” (whatever that means, for even its advocates seem to have no coherent idea) is probably responsible for the modish elevation of feelings over evidence. One of the many cartoons that adorn this book shows a child who thinks seven times five equals 75 and says, “It may be wrong but it’s how I feel.” The normally genial Pinker takes the gloves off in the penultimate chapter, “What’s Wrong with People,” and he can be devastating. Here, for instance, he laments the sinking public trust in universities and attributes it to the universities’ suffocating …

monoculture, with its punishment of students and professors who question dogmas on gender, race, culture, genetics, colonialism, and sexual identity and orientation. Universities have turned themselves into laughingstocks for their assaults on commonsense (as when a professor was recently suspended for mentioning the Chinese pause word ne ga because it reminded some students of the racial slur).

It’s not just assaults on common sense. There’s the wanton debauching of the English language. To clarify the distinction between logical and empirical truths, Pinker says, “To determine whether ‘All bachelors are unmarried’ is true, you just need to know what the words mean (replacing bachelor with the phrase ‘male AND adult AND NOT married’) and check the truth table. But to determine whether ‘All swans are white’ is true, you have to get out of your armchair and look.”

What could be clearer? But is Pinker bulletproof against an oppressed minority of married males who identify as bachelors—those who feel, deep down that, though married they truly are bachelors—and Pinker’s logical point threatens their very existence. Pinker is worried that this kind of nonsense threatens the reputation of universities. I worry that every campus news story of this kind (and there are many) nets another fat bushel of votes for Donald Trump and his odious acolytes.

The middle chapters amount to a patient instruction manual on how to think rationally. I wouldn’t call it a primer; this is state of the art. After the rather demanding chapter on logic, we have a deep discussion of the theory of probability. Pinker is a committed devotee of Bayesian reasoning, and here I learned a lot. It recurs as a leitmotif throughout the book. A chapter expounds the economists’ notion of rational choice and risk. And of course game theory. The important distinction between correlation and causation gets a whole chapter. The rooster always crows at dawn but which, if either, is cause and which effect? The only sure way to demonstrate a causal relationship is an experimental intervention. Artificially administer the putative cause at random and see whether the putative effect follows. Playing a recorded cock-a-doodle-doo in the middle of the night to see if the sun promptly appears is a stretch, but you get the idea. Of course, you must do it many times and at random. Unfortunately, such experimental intervention is often not possible, especially in studies of human behavior, and Pinker devotes much of the chapter to statistical tricks, which are the next best thing. This leads him to lay out the powerful technique of multiple regression.

There’s a chapter on Signal Detection Theory, which has long been a favorite topic of mine, and I was glad his lucid treatment graduated from the classical zone of blips on radar screens to such human-interest topics as decisions by juries. What might courts do to increase d-prime (mathematical jargon for the discriminating power of evidence)? Pinker quotes the disturbing findings of Elizabeth Loftus and others on the unreliability of eye-witness testimony. DNA evidence is a boon by comparison. If properly used, DNA identification is equivalent to picking out one man from an identity parade of millions—instead of “at least eight” (as is the rule in English law, for example). Even DNA evidence is vulnerable to accidental contamination or mislabeling of samples, but such human errors can be overcome.

How should beta (the decision criterion) be set in a court of law? Should we tolerate condemning one innocent person while letting ten guilty ones go free? What is meant by “reasonable doubt”? I have long thought that if the verdict really is beyond reasonable doubt, there should be no nail-biting tension when the jury returns. It should be obvious to everyone in the court who has heard the same evidence as the jury. Isn’t “obvious” pretty much what we mean by “beyond reasonable doubt”? And “beyond reasonable doubt” should also mean that if two juries sat through the same trial and then retired to their separate jury rooms, they should always return the same verdict. But would they? You cannot be serious! Think O.J. Simpson. Might the obviousness of “beyond reasonable doubt” even obviate the need for hot-shot lawyers?

“We evolved not as intuitive scientists but as intuitive lawyers.” We evolutionary biologists can well believe that, being members of a highly social species, our ancestors were selected for their ability to outwit, out-general, and out-persuade rivals for mates and economic goods. The same competitive social milieu might have selected our ancestors for their ability to set up alliances—rival gangs. Perhaps this accounts for the “myside bias,” which today is attested by ingenious, if depressing, psychological experiments.

Perhaps the most depressing of these experiments presented Republicans and Democrats with (made-up) data superficially purporting to show that gun control decreases crime. In fact, a critical analysis of the data would show the opposite: gun control leads to an increase in crime (remember this was made-up data!). Only numerate respondents saw through the superficial conclusion. But the interesting fact was that it was numerate Republicans who mostly achieved the necessary perspicacity—supporting a conclusion in favor of their political bias. Numerate Democrats tended to rest satisfied with the superficial conclusion that supported their bias. But here’s the kicker: Republicans and Democrats were also presented with the identical made-up data but with the headings flipped. Now the data superficially supported the gun lobby, and again you had to be numerate to spot that they actually led to the opposite conclusion. Now it was the numerate Democrats who penetrated to the correct inference, the one that supported their bias. Finally, when the experiment was repeated with the fictitious issue being not gun control but an anodyne one where neither tribe had a dog in the fight, there was no difference between Republicans and Democrats. The depressing conclusion is that we humans believe whatever supports our team, irrespective of the evidence. Indeed, I find it depressing that we even have teams, rather than an abiding loyalty to evidence.

The last two chapters, “What’s Wrong with People?” and “Why Rationality Matters,” are the ones that I most copiously annotated. If I were to do justice to them, I’d have to quote almost every sentence; they’re all so good. I’ll just close by contrasting the malign gullibility of “Stop the Steal” QAnon believers, on the one hand, with the disinterested rationality of trained scientists and of Chapter One’s Kalahari hunter-gatherers, on the other. How is it possible that the same species includes members capable of measuring the infinitesimal shivering of spacetime itself and others capable of believing that Hillary Clinton runs a pedophile ring from the premises of the Comet Ping Pong pizza restaurant?