It’s a busy time for the world’s most famous defender of Enlightenment liberalism. The Harvard psychologist Steven Pinker, a co-founder of his university’s Council on Academic Freedom, has spent much of the past decade fighting threats to academic free speech from cancel culture and the “woke” left.
Now, American universities find themselves challenged by the Trump administration’s campaign to defund research. “I’m kind of pinned between those two trends,” Pinker tells me when we meet in central London. Which poses the greater threat? There’s no question. “Trump has an army … whereas the Department of Romance Languages — there’s only so much damage they can do.”
At 71, many professors might be content to wind down. Pinker’s willingness to say what he thinks (rare among academics) and his skill at elucidating complex ideas (rarer) means he is in the midst of what looks, from the outside, like the busiest moment of his career. Over the past few years he has written tirelessly in defense of science and academic freedom. And his expertise in cognitive science and linguistics means he is much sought out for his views on AI. He is a prolific podcast guest. Now there is a new book, When Everyone Knows That Everyone Knows …, a characteristically lucid account of the fiendish and unfamiliar psychological concept of “common knowledge”.
Pinker originally rose to prominence as a superlatively gifted science writer; he is the author of serious but accessible books on linguistics and cognitive science, of which the best known are probably The Language Instinct (1994) and How the Mind Works (1997). As a popularizer of scientific ideas, his only real rival is his friend the biologist Richard Dawkins. The epithet “controversial Canadian psychologist” began to attach itself to Pinker’s name after the publication in 2002 of The Blank Slate, which debunks the fashionable idea (beloved on the progressive left) that human nature is a cultural construct, owing little or nothing to our genetic inheritance.
More sacred cows were prodded in Enlightenment Now (2018), a polemical defense of the Enlightenment values of science and reason, which, Pinker argues, have set humanity on a centuries-long upward trajectory toward peace, health and prosperity. To his admirers Pinker is a force of reason and rational optimism. To his detractors he is the politically incorrect reincarnation of Voltaire’s Dr Pangloss.
As is often the case with public figures considered controversial, he doesn’t come over that way in person. Pinker is polite, wry and softly spoken in his smart suit and blue shirt. I’m glad to see his luxuriant mane of curling white hair (which bears an apt resemblance to the periwig of an Enlightenment philosophe) is going strong.
Fittingly for the author of a book titled Rationality, Pinker has a conversational habit of making an apparently casual point, then citing a social psychology experiment to back it up, sometimes going on to evaluate the quality of the evidence he has just cited. When I ask Pinker whether his students use AI to write their essays he tells me Harvard students sign a pledge promising not to let it write their papers for them. “There used to be research that signing a pledge prior to filling out a form makes you more honest,” he says smiling, “but sadly, that finding turned out to be fraudulent.”
“Common knowledge”, the subject of his new book, is a psychological concept that occupies a strange zone between the blindingly obvious and the mind-blankingly counterintuitive. Used technically by psychologists, “common knowledge” refers not just to a condition in which “everyone knows” something to be true, but to when everyone knows something is true and also that everyone else knows it. In the story The Emperor’s New Clothes, everyone can see the emperor is naked. It is only when the little boy points his finger and blurts out the truth that everyone knows that everyone else knows. It’s then that the crowd falls about laughing.
Common knowledge sounds simple, but Pinker deftly guides readers down labyrinths of “recursive mentalizing” — ie what I’m thinking about what you’re thinking about what I’m thinking about what you’re thinking, and so on ad infinitum. Consider the layers of face-saving ambiguity involved when a man propositions a woman by inviting her upstairs for coffee after a date. She might think: “I know that I’ve just been propositioned, but does he know that I know? Maybe he thinks I’m naive and that I’m just turning down coffee.” You might then ask: “Does he know that she knows that he knows?” Perhaps she thinks he’s just dense. Common knowledge underpins the “innuendos and hints” on which human lives and relationships depend.
Common knowledge is also crucial to enforcing social norms that exist “only insofar as everyone knows they exist and knows that everyone else knows it”. Historically, criminals were punished publicly so that “onlookers could see one another seeing the spectacle”. This made it a powerful mechanism of social control. Pinker argues that “social media has equipped a billion people to step into the role of common-knowledge norm enforcers” capable of inflicting public punishments on others “at a low cost to themselves”. This proliferation of common knowledge may be why the past decade of American life has seemed “uniquely stupid”.
I ask Pinker whether we are witnessing an anti-rationality backlash. He suggests what we’re seeing is “greater inequality in irrationality”. Sport has never been more rational: nowadays in America every team has a statistician. Also, “there’s more evidence-based medicine” than before. But “at the same time as that, irrationality has gotten absolutely entrenched at the highest levels of power in the United States”.
He thinks the roots of the problem may be partly traced to “the politicization of science”. During Covid there were “hundreds of public health experts saying that it’s OK to go out in Black Lives Matter protests” because “the benefits of social justice outweigh the costs of spreading Covid”. They would not have said the same had it been “say, a Maga rally”. He cites a number of similar examples, including an academic journal that promised to “consult members of indigenous minority groups before deciding whether to accept scientific papers” and a science magazine that endorsed Hillary Clinton for president. Such things “erode the credibility of science”. He says: “If science as an institution brands itself as on the political left, it should be prepared to alienate and maybe kiss off the American right.”
Not, of course, that he has sympathy for Maga anti-science sentiment. He’s perturbed that the Trump administration is taking steps to bar Harvard from receiving federal research funding. “Harvard’s entire science, engineering, medical, public health departments depend to a large extent on federal funding,” he says. “The damage could be catastrophic.”
Would he go into academia were he starting his career over again? He says he would make the same choices, but that some of his students are definitely put off. One has just taken a job at a think tank because he’s “afraid that there’s no future for him because of his opinions”.
Another challenge facing universities is AI, with which student essays are supposedly riddled. Is it a problem? He says that it is in so far as writing is important for imparting critical thinking, because putting ideas on paper helps us to give our thoughts “greater precision”. He doesn’t use AI much himself since he is wary of its “tendency to hallucinate” (make stuff up), which is “built into the architecture” of large language models (LLMs) like ChatGPT. For this reason he’s skeptical of the wilder predictions of an AI-powered transformation of white-collar work. LLMs are unhelpfully prone to producing “third rate, barely passable, superficially plausible output that humans have to either weed out or perfect”.
What about reports that ubiquitous digital distractions are affecting students’ ability to learn and think? “It is a serious problem,” he says. One of his Harvard psychology lectures begins with a “mind-blowing 3D demo playing on the screen”. It’s only three minutes, but his students are “on their phones, they’re not even looking up”.
Although he’s aware that “every aging professor thinks the kids are getting worse” he suggests that smartphones are probably “not disconnected from declines in academic performance”. Characteristically, he has data. Every year, he gives a final exam to his students. “I plotted performance in the multiple choice test since 2004. That’s a span of 21 years and performance has gone down 10 percentage points from 85 per cent correct to 75 per cent correct.”
But, as Pinker is careful to point out, “with any new technology, it’s very hard to know when it’s introduced what the workarounds will be”. Let’s hope we find them. We need the next generation of Steven Pinkers. In an age when the term “public intellectual” seems to stick to anyone capable of ranting into a microphone, When Everyone Knows That Everyone Knows … is a welcome reminder of what the real thing looks like.
James Marriott is a columnist at The Times of London, covering society, ideas, and culture
