“There’s trouble at t’lab”
Stuart Ritchie’s ‘Science Fictions’ reveals a scholar committed not only to his own discipline but to the wider principles underlying all intellectual endeavour
If you’re a Tory (and if you’re not, you probably laughed at the photographs), you’ll recall Theresa May waddling onto various stages at Conference like she’d suddenly grown a pair of large brass balls. She was not alone in this — the Cabinet joined in, with Sajid Javid looking particularly ridiculous. After the laughter died down, it became apparent that the oldest and most successful political party in history had been taken in by a bit of junk science: power posing.
Both Cynical Theories and Science Fictions deserve to be enormous bestsellers
“Power posing”, science told us, caused testosterone to surge, cortisol to plunge and people who did the various stances to feel more confident. There were books, TED-talks, and studies to prove this. Except science did no such thing. The original research (in 2010) was based on just 42 people, and when other scientists came to consider its claims, the findings that persuaded sundry politicians to walk, well, a bit like ducks, failed to appear. Apart from amusing the Great British Public, the collective national guffaw at power posing introduced many of us to something that’s been bubbling away for the best part of a decade: the replication crisis.
2020’s coronavirus misadventures had a ruinous effect on book publishing, and to my mind the most unfortunate victims were Helen Pluckrose and James Lindsay’s Cynical Theories: How Universities Made Everything about Race, Gender, and Identity – and Why This Harms Everybody and Stuart Ritchie’s Science Fictions: Exposing Fraud, Bias, Negligence and Hype in Science.
Both Cynical Theories and Science Fictions deserve to be enormous bestsellers and in a non-pandemic year they would have been. Taken together, the two books dissect not only what afflicts universities throughout much of the developed world, but convey the deleterious effects of muddy thinking and cut-rate reasoning on politics, policy, and the wider life of the mind. Pluckrose and Lindsay cover off the humanities, while Ritchie focusses on STEM. There is an overlap where both address the social sciences: Pluckrose and Lindsay consider less numbers-oriented disciplines like sociology and anthropology. Ritchie trains his aim on fields that depend for much of their salience on statistics, like psychology and economics. Imagine, then, a Venn Diagram with “STEM/Science Fictions” in one circle, “Humanities/Cynical Theories” in the other, and “social science” in the shaded region.
I gave Cynical Theories the prominence it deserved for The Critic in August. Today, allow me to give Science Fictionsits due.
Science Fictions is a work of great moral seriousness and its arguments need to be digested
In telling the hilarious power posing story (and many others like it), Ritchie is revealed not only to be a dab hand at narrative exposition, but also a good-humoured Virgil guiding the reader’s Dante through a mire of dishonest, biased, negligent, overhyped science and the perverse incentives that create it. Although Ritchie (I think rightly) treats his own discipline, psychology, as the canary in the replication coalmine, he ably conveys the extent to which something that first emerged there is an unfolding catastrophe across the sciences. That is, published and sometimes famed results simply disappear in later attempts to make them happen again.
“How many times has your doctor prescribed you a drug or other treatment that rests on flawed evidence?” Ritchie asks at one point. “How many times have you changed your diet, your purchasing habits, or some other aspect of your lifestyle on the basis of a scientific study, only for the evidence to be completely overturned by a new study a few months later? How many times have politicians made laws or policies that directly impact people’s lives, citing science that won’t stand up to scrutiny? In each case, the answer is: a lot more than you’d like to think.”
Not all of Ritchie’s stories are funny, either. His section on fraud is eye-popping and soberly told, because it has to be. Among others, it introduces Paolo Macchiarini, whose situation I thought I knew thanks to it being discussed in publications directed at medical negligence lawyers.
Macchiarini was a thoracic surgeon who claimed to be able to transplant tracheas — something I didn’t realise is exceptionally hard to do, even when compared to formidably difficult heart and lung transplants. He published a number of papers touting his success at “seeding” artificial tracheas with some of the recipient’s stem cells to forestall the immune system rejection that makes organ transplantation so difficult.
While Macchiarini’s papers were lighting up the medical and surgical literature and he was being hailed a hero at Sweden’s famed Karolinska Institute (where he worked), patients who received trachea transplants using his methods were dying. Even worse, it not only took years for the scandal to come to light but both Karolinska and The Lancet —which had published some of his work — continued to run interference when outsiders (including journalists) attempted to investigate. It was only in 2018 that The Lancet retracted his papers and formally stated he was guilty of scientific misconduct.
Science is one of humanity’s greatest developments. Academia, on the other hand, is not
Even when it isn’t killing people, bad science does bad things (in addition to undermining public trust in science). The austerity policies introduced in many countries after the Global Financial Crisis in 2008-9 were based in part on what appeared to be an unusually compelling economics paper showing that when a country’s debt-to-GDP ratio rose above 90 per cent, growth stalled. Except it didn’t: the authors had made a mistake in their Excel spreadsheet (a phenomenon familiar to us thanks to Public Health England’s recent Microsoft misadventures) and accidentally omitted a number of countries’ debts. Two of the countries left out were Canada and Australia. The former was hardly dented by the global financial crisis and the latter emerged wholly unscathed. Fixing the Excel typo blew up the debt-to-GDP finding and grievously wounded the case for austerity. Too late, of course, for Greece, which was crippled by it.
The Excel-impugned economics paper is in the book’s “negligence” section. At this point — for all that stories of medical fraud leading to multiple deaths make one despair at science — Ritchie is clear that negligence, bias, and hype are far more common and, in the long run, more serious. This is particularly the case when it comes to bias. Notably (and unlike the humanities, which Cynical Theories reveals has been poisoned by left-wing ideology), Science Fictions suggests that bias in STEM is only occasionally political.
Much of the time, the partiality Ritchie identifies is that of scientists who desperately want something to be true. To make it true, they then torture their data until it confesses, indulging in a practice known as p-hacking. P-hacking involves using statistical tricks to make results look bigger (“more significant”) than they really are. When coupled with publication bias, science’s equivalent of Procrustes’ Bed, the consequence is that vast numbers of even major studies are not worth the pixels on which they typically appear — studies that don’t “fit” are simply chopped off at the socks. That said, when politics in science does rear its head, it’s just as ugly as the stuff Pluckrose and Lindsay discuss in Cynical Theories. I’m afraid Stephen Jay Gould comes out of Science Fictions with only some of his reputation intact.
Ritchie also describes a phenomenon where people often assume a finding is true because it’s been peer-reviewed, when in reality frequent citation of previous publications that fail to replicate misleads individuals, groups, and governments into believing there is proof for something. The number of papers still receiving positive citations and commentary despite being formally retracted by the journals in which they were published is staggering and to my mind probably the most alarming part of Ritchie’s book. This is how untrue claims are first accepted and then fed into policy development.
To protect science, the academy needs fixing
Not a few reviewers have blown Ritchie off because he has a light, breezy style. Ignore that — or only giggle when you really can’t help it. Science Fictions is a work of great moral seriousness and its arguments need to be digested. Those who’ve criticised Ritchie for focussing on problems in science since 1960 — often by observing that there has been scientific fraud, bias, hype, and negligence in every society with some grasp of what we now call the scientific method going back to classical antiquity — don’t get one of his main arguments. Which is, there is so much more of all of those things now not only because we have more science and scientists, but because higher education has grown like kudzu, generating perverse incentives that mean even sound and ethical individuals are tempted to engage in bias, hype, and negligence if not actual fraud purely to get and hold a job.
Science’s garden is dire need of weeding and failure to do so will lead to people resenting their tax moneys being wasted so egregiously and who do want to burn it all down as a result. And it isn’t only tax money going up in smoke. Ritchie’s discussion of badly designed studies using animals — where animals are not only hurt or killed but no study is ever published — is distressing to read and suggests there’s widespread callousness as well as laziness in the academy. A number of conservative governments globally are seriously annoyed with the academic humanities and are quite deliberately using Covid-19 as cover to pull or cut funding. The threat to academic science — if it fails to clean up its act — is just as real, and probably bipartisan.
Science is one of humanity’s greatest developments. Academia, on the other hand, is not. Ritchie gets this intimately: anyone who thinks he also wants to “burn it all down” has entirely missed the point. Science Fictions reveals a scholar committed not only to his own discipline but to the wider principles underlying all intellectual endeavour. However, to protect science, the academy needs fixing. Reading this book is a good place to start.
Enjoying The Critic online? It's even better in print
Try five issues of Britain’s newest magazine for £10Subscribe