The world out there can often seem as though it is hurtling to hell in a handcart: people are refusing safe vaccines for a dangerous disease, extreme weather events caused by global heating are on TV nightly, billionaires are shooting themselves into the stratosphere in penis-shaped spacecraft while record numbers of the precariously employed rely on food banks. Looked at from this perspective, humanity as a whole doesn’t seem very rational. Hence why, surveying the idiocies of his own age, Jonathan Swift amended Aristotle’s definition of humans as “the rational animal” to his own sardonic formulation animal rationis capax – the animal capable of rationality.
How, though, should we become more capable? Most of the time, thinking sounds like hard work, but add “smart” to the front and it sounds more attractive: hipsterishly mid-Atlantic, vaguely technological (like “smartphone”), and with an implied promise of some handy trick or shortcut. A person who is smart – etymologically “sharp” or “stinging” – rather than merely thoughtful or intelligent is someone endowed with a certain practical cunning, not a dweller in ivory towers. Hence the rise in publishing of the “smart thinking” book, an elevated species of self-help for the aspiring ratiocinator.
The yin and yang of the modern smart thinking genre are represented by two of its foundational bestsellers: Malcolm Gladwell’s Blink (2005) and Daniel Kahneman’s Thinking, Fast and Slow (2011). The former celebrates the unconscious information processing that enables fast, “gut” decisions to be sometimes more accurate than those involving careful reflection. Kahneman, by contrast, emphasises the fallibility of snap judgments, detailing the work he did with Amos Tversky in behavioural economics, which suggests that instinctive, rapid cognition is prey to a host of severe errors. These are the now familiar “cognitive biases”. One is the availability bias: if there has recently been a terrorist incident in the news, we are likely to think that terrorism is more common than it really is. Another is the anchoring effect, where arbitrary numbers affect our subsequent estimates of something completely different. In one of Kahneman and Tversky’s experiments, participants were told to spin a wheel that gave a number between 0 and 100, and then asked how many African countries were members of the UN. Those who had spun a high number on the wheel gave a higher answer to the question.
A decade ago, the fashion was to be pessimistic about the prospects of improving our thinking, and even about the value of thinking at all. Another bestseller about cognitive biases was entitled You Are Not So Smart (David McRaney, 2011), while, in The Righteous Mind (2012), the social psychologist Jonathan Haidt claimed that the idea rationality is our most noble capacity is a “delusion”, and that “worship of reason” is “an example of faith in something that does not exist”. (If so, one had better give up writing and reading books.)
Such outright nihilism is unusual, though. One of the ancient maxims inscribed in the temple of Apollo at Delphi was gnothi seauton – “know thyself” – and most such books, including Blink and Thinking, Fast and Slow, claim to educate the reader in (rational) theories of how we think, in order to help us think better. Kahneman himself is sceptical about our ability to reliably “de-bias” our own judgments, but the hard sell of Gladwell’s Blink is that, where it is fallible, the gut instinct can be retrained: “We can teach ourselves,” he promises, “to make better snap judgments.”
“De-biasing”, indeed, has become an industry, with one survey of how unconscious bias leads to social harms, and how it may be combated, offered in Pragya Agarwal’s Sway, published last year. Jessica Nordell’s new book The End of Bias: How We Change Our Minds, while not a smart thinking book in the classic self-help sense, is an intriguing survey of strategies to combat racial and gender bias in particular, including mindfulness and meditation training for police officers, the use of checklists in hospitals, and the diversification of university faculty to give prospective students visible role models.
As all such books must do, The End of Bias references the classic work by Kahneman and Tversky, as well as another landmark of the genre, Cass Sunstein and Richard Thaler’s Nudge (2008). The problem addressed by this latter book is: what should smart people do about the fact other people are less smart than they are? Rather than teach them to be smart, we can instead trick them into doing what we consider is in their best interests by altering the “choice architecture” of their environment so that their cognitive biases will lead them to the correct decision. Hence the interventions of “nudge politics”, which include placing healthy foods at eye level in the supermarket or making employee pension savings accounts opt-out rather than opt-in.
So are we irredeemably dumb animals to be shepherded into good behaviour by the subterfuge of nudging, or are we perfectible thinkers, if only we read the right handbook? These dynamics are still working themselves out in the latest crop of smart thinking books, and not only when they are by the same authors. Kahneman recently teamed up with Sunstein and Olivier Sibony, a business professor, to write Noise: A Flaw in Human Judgment, which purports to have discovered a new weakness in our thinking – “noise” being inconsistency in judgments – that can, luckily, be rationally reduced. This may be done with what has become the standard curriculum of smart thinking books: be aware of the cognitive biases, do not commit the “base-rate fallacy” (where you ignore the wider prevalence of something in the general population), scrutinise the sources of your information, and so on.
Two other new smart thinking books, meanwhile, are considerably more optimistic than used to be mainstream. The mathematician Marcus du Sautoy’s enjoyably clever Thinking Better: The Art of the Shortcut emphasises dutifully at the beginning that “we have limited mental capacity to navigate complex problems” and that “most of the heuristics [rules of thumb] that humans use lead to bad judgments and biased decisions” – but the good news, the book shows, is that mathematics is not like that. Maths is the art of the successful shortcut, as he goes on to demonstrate with vividly illustrated chapters about the real-world applications of algebra, geometry, probability theory, and so forth. Probability theory, naturally, is useful if you visit casinos (it was invented to solve gamblers’ arguments in the 17th century), geometry enabled the ancient Greek Eratosthenes to measure the circumference of the Earth to a surprising degree of accuracy without walking all the way around it, and calculus is crucial to the design of roller coasters.
He also explains the special flavour of probability theory called Bayesian, after the 18th-century English minister Thomas Bayes. The insight here is that when estimating the likelihood of something, you have to take into account relevant prior information that is already known. Say there is a medical test that is 90% accurate at detecting cancer, and your test comes back positive. “Most people will be freaked out,” Du Sautoy notes, but “it is important to have the extra data that only 1 in 100 patients is likely to have cancer”. Out of those 100 people, though, another 10 will get false positive results from the test, since it’s only 90 % accurate. So we have 11 people with positive results, but only one person who is really ill. So all you should conclude from a single positive test is: “There is only a 1 in 11 chance that actually you are the one with cancer.” The information about the background prevalence of the disease is an example of what is called a “base rate”, and our failure to consider it is yet another common cognitive bias, christened the “base-rate fallacy”.
Bayes also features in Steven Pinker’s noble and slightly self-satisfied defence of reason, Rationality: What It Is, Why It Seems Scarce, Why It Matters, rejects the claim of deep human irrationality outright. “As a cognitive scientist I cannot accept the cynical view that the human brain is a basket of delusions,” he announces. “A list of the ways in which we’re stupid can’t explain why we’re so smart” – smart enough, say, to invent a vaccine for a novel virus in a matter of months, or to put a rover on Mars.
Pinker follows the arguments of dissident researchers such as Ralph Hertwig and Gerd Gigerenzer, who argued at the time that Kahneman and Tversky’s results on cognitive bias did not always entail a diagnosis of irrationality. He also emphasises that rationality is not culturally or historically specific: modern hunter-gatherers in the Kalahari, he shows, track their prey using exactly the same methods of logical inference employed by celebrated Harvard psychology professors. The book runs through decent explanations of Bayesian probability theory (as applied to gambling or medical test results), correlation and causation, and so forth, before veering a bit off-piste to denounce the “suffocating leftwing monoculture” of universities.
But who are the “we” who read the descriptions of ourselves in such books? Publishers of general nonfiction like their products, if possible, to address a “soft business” demographic, which is why so many smart thinking books offer their wares as a way to get ahead in employment. Arguably the godfather of the genre is the recently deceased psychologist Edward de Bono, who first coined the phrase “lateral thinking” in the 1960s and subsequently developed a range of franchised corporate workshops on “de Bono’s Lateral Thinking™”, “de Bono’s Power of Perception™”, “de Bono’s Six Thinking Hats™”, and so forth.
In an age of neoliberalism, reason is pictured as a private possession, a trump card of one-upmanship. Kahneman, Sunstein and Sibony’s Noise is explicitly addressed to people troubled by such questions as “How much will my product sell?” or “How did my assistant perform last year”? Du Sautoy, mindful of the genre’s requirements, even suggests that mathematical thinking might help you get rich by spotting patterns in share prices, though to his credit he doesn’t really seem to believe it. Pinker, meanwhile, extols the rationality of economic “rational choice theory”, introducing it simply as a theory of mathematics that has “no direct implications for how members of our species think and choose” – and yet by the end of the very same paragraph he claims that “it can be a source of profound life lessons”. In the atomised marketplace of ideas, life lessons are a product that you can buy with smart thinking. If we are really concerned about rationality, mind you, every introduction to game theory – the study of rational decisions, the outcomes of which are also dependent on the rational decisions of other people – should point out, as Pinker omits to do, that John von Neumann, the discipline’s co-founder, urged President Truman to pre-emptively nuke the USSR in 1950. Luckily, this life lesson was not acted on.
Du Sautoy’s Thinking Better is more honest than most in admitting the limits of its applicability. He acknowledges that some things just can’t be life-hacked: there are no short cuts to learning to play the cello well, a truth he goes to the trouble of having confirmed to him by a professional cellist. Similarly, Pinker offers a subtle undermining of popular media in his arch observation that the word “surprising” (in headlines or books offering surprising truths) really means “unlikely to be true”. By contrast, he says, we need “a healthy respect for the boring”. And yet publishers won’t pay advances for boring books. So here we are.
A deeper problem with the smart thinking genre as a whole is that it commits the sin identified half a century ago by the Frankfurt School philosopher Max Horkheimer, in his Critique of Instrumental Reason (1967). “Reason is considered to come into its own,” he lamented, “when it rejects any status as an absolute (‘reason’ in the intensified sense of the word) and accepts itself simply as a tool.” Fifty-four years on, this idea – that reason is “simply a tool” – is so internalised as to appear the most basic of common sense. What is rationality? Why, says Pinker, simply “a kit of cognitive tools that can attain particular goals in particular worlds”. The inevitable upshot of such an impoverished view of reason, Horkheimer thought, would be “the automatising of society and human behaviour”, and a “surrender to blind individual and national egoism”. Present readers may judge for themselves whether he turned out to be right.
The instrumentalisation of reason, in Horkheimer’s view, went hand in hand with society becoming ever more “irrational” in the true sense. “Nowhere does the union of progress and irrationality show up so clearly as in the continued existence of poverty and care and the fear of distress and dismal old age, and in the condition of brutal prisons and asylums in countries with highly developed industry,” he wrote. What this gestures at is what so few modern smart thinking books acknowledge: that rationality is not a private possession but a public institution. Reasoning is fundamentally social, one trivial proof of which is the fact that we all get most of our reliable knowledge about the world from authorities (scientific and otherwise) without doing our own personal experiments.
It’s Du Sautoy, in the end, who provides the wisest commentary on the limits of instrumental reason by way of a quotation from the Swiss mathematician Leonhard Euler, who wrote in 1808: “It is not knowledge, but the act of learning, not possession but the act of getting there, which grants the greatest enjoyment.” This ideal represents the opposite of what is sold to us under the smart thinking rubric: it is thinking that cannot be instrumentalised and converted easily into status or profit.
Is there, for all that, reason to have hope for humanity? “Our picture of the future need not be a bot tweeting fake news for ever,” Steven Pinker writes cheerily. But we’ll need more than a library of smart thinking books to avoid that fate.
from Hacker News https://ift.tt/3mkwDu4
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.