On today's episode of "Speaking of Bitcoin," correspondent George Frankly explores the expertise paradox through the lens of two very different historical figures.
Hello there – I'm George Frankly and I'm going to take a look at how even the best and brightest people can make truly stupid decisions and terrible predictions, and what we can learn from them. This is "Dare to Be Stupid."
This time on "Dare to Be Stupid," “The Expert Incentive,” or “God Save the King of Kong.”
I’m not looking forward to this. I loathe misinformation. I loathe innumeracy. And before anybody asks what the word innumeracy means let me say I also loathe illiteracy. The world needs experts and experienced professionals in positions of power, now more than ever. I firmly believe that.
So it honestly sucks that I’ve decided to sit here today and tell you that expertise is dangerous.
Experts are a complicated concept, and despite agreeing on the broad strokes of the word we all have granular disagreements in what being an “expert” actually means. I want to get into the nitty-gritty of where expertise goes right and wrong by talking about two sides of the same coin, two closely intertwined experts in their respective fields:
As a dedicated fan of both men, it’s hard to know where to begin. Semmelweis was an expert that pioneered the earliest concepts of germ theory, and Mitchell is an expert that cheats at the 1981 arcade game Donkey Kong for money.
The key difference is that Semmelweis’s fellow experts insisted he was a fraud, and Mitchell’s contemporaries insisted for years that he wasn’t.
Ignaz Semmelweis studied law, science and medicine at the University of Vienna and made his mark on history working at the obstetrics clinic in Vienna General Hospital. There, he was frustrated by the high rates of maternal mortality in the clinic. The death of mothers by post-childbirth fever was uncomfortably common, and he wanted answers. Infection and germ theory were not yet understood concepts, so all he had was deductive reasoning and experimental design – that thing your middle school teachers called “the scientific method.”
The immediate observation that spurred him forward was the most confusing one. The mortality rate for new mothers was as high as 15% in the practicing doctors’ ward, but in the next ward over, which was staffed by traditional midwives instead of university-educated doctors, the average mortality rate was under 5%. The rate of contracting fatal childbed fever was three times higher in the clinic run by the elite doctors than in the clinic run by the supposedly unprofessional midwives. He was certain that if he could isolate the cause of the lower rates, he could adapt it into the other clinic and save lives. I’ll give you a hint: He was right. But we’ll get to that.
Billy Mitchell is an icon. An icon of what varies from person to person, but he is nonetheless iconic. For over 25 years he’s rarely been seen without a sharp suit, loud American flag necktie and a magnificently coiffed, flowing mane of hair. He was one of the earliest faces of video game world records – a massive hobbyist sport of competing for the fastest times and highest scores in arcade and home video games.
Mitchell rose to fame with his record of the world’s first perfect Pac-Man score in 1999 and further for his world record high score in Donkey Kong. The 2007 documentary "The King of Kong: A Fistful of Quarters" chronicled his dramatic victory and cemented his public image as a fierce competitor, and a bit of a cinematic villain.
His personal ties to the company that officiated and validated the scores allowed him to skirt a lot of the scrutiny that other competitors faced. Whereas most world records had to be achieved in a public exhibition on inspected hardware, Mitchell was able to turn in a fuzzy VHS tape of his record Donkey Kong run at home and had it accepted on the spot. The suspicion soon boiled over, and by 2018 multiple detailed technical analyses found that Billy Mitchell had cheated.
About 170 years previously, Dr. Semmelweis had catalogued several differences and tested their correlations over many long months. Midwives delivered with women on their sides – he had doctors move women to their sides. No effect. Patients in the doctors’ ward were regularly visited by priests with loud bell-ringing attendants. He had them ditch the bells. No effect.
But then, inspiration struck. One of his co-workers died. Wait, that sounds terrible. No wait, it is terrible. But, grim or not, it gave him a lead: A fellow research doctor had accidentally lacerated his own hand while performing an autopsy, and succumbed to rising temperature and death identical to the stages of childbed fever. The link to the autopsy was a bit circumstantial, but Semmelweis noted that nearly all of the doctors in his ward did routine autopsy research – the midwives never handled cadavers.
He had a frankly absurd theory. He suspected there were “fine cadaverous particles” (in his words) being transmitted to vulnerable women and causing the illness. Midwives were rarely exposed to corpses and routinely rinsed their hands with hot water and soap, so he adopted a similar process: Doctors in his ward would regularly rinse their hands and tools in a chlorinated lime solution – weak bleach, essentially. He guessed – very luckily – that the chlorine solution’s ability to cleanse strong odors meant it might remove his mystery particles. He had discovered germs and then disinfectants, long before either could be fully explained.
The maternal mortality rate in his ward eventually dropped under 2%, and at a few points even went entire months with no deaths. He brought his method to his next hospital posting, where rates dropped from 10% to less than 1%. He soon wrote a paper about his findings and began to distribute it across the region.
It was widely and aggressively rejected. Semmelweis’s core thesis of “please wash your goddamn hands” went against all of the expert consensus. His insinuation that it was doctors – academics and experts every one of them – doctors who were spreading childbed fever was considered insulting, hostile and an affront to their authority. His hand-washing solution was seen as little more than magical thinking without scientific merit, and the medical establishment shunned him. He eventually died in a mental institution of complications from a hand injury, the exact same fever he had railed against. It wasn’t until decades later that he would be vindicated by Louis Pasteur’s seminal work on germ theory, and the history books have proven kinder to him than, well, kinder than history itself was.
In the 21st century, Billy Mitchell faced the opposite problem. A small but vocal group of analysts and programming experts had found ultra-fine discrepancies in his world record video. At first, coding experts found that Mitchell’s luck was supernaturally high – the random elements of the game were suspiciously generous during the recording. Soon they found that split-second variations in how the ladders and girders of Donkey Kong’s stages loaded onscreen didn’t match the display of real arcade hardware; in fact, it perfectly matched the behavior of PC emulator software. Mitchell had essentially pulled off his miracle run on a home computer simulation of the game, with access to limitless modifications and instant do-overs.
But Billy Mitchell was unbowed. The highly technical evidence struck many as unconvincing. More than that, Mitchell and his fans touted a bigger point: He wouldn’t need to cheat. Billy Mitchell had numerous public performances of extremely good Donkey Kong runs. He knew all the ins and outs of the game. Billy Mitchell was, indisputably, an expert at the game of Donkey Kong. Why would an expert bother to cheat?
Ignaz Semmelweis was called a fraud because he defied the experts. Billy Mitchell was called the real deal because he was an expert. By most definitions, all these experts are experts. Why did expertise choose the wrong side of history both times?
That’s easy. They’re all people. And being an expert doesn’t cure you of common reasoning errors. In fact, it can teach you all-new ones.
The doctors weren’t just guided by simple egotism, although that was definitely a factor. No, they fell into something that sociologists have begun to call “the expertise paradox.” Specialists in a field – including the very best in their fields – inevitably develop tunnel vision. Hyper-specialization in a field or an industry begins to close them off to outside knowledge. To a man with a hammer, every problem is a nail: To a man with a prestigious medical degree in 1830, microbiology looks like magical bulls**t.
Innovation rarely grows from within a rigid institution, much less its tenured experts. Hell, Donkey Kong, a groundbreaking arcade game that revitalized and revolutionized video games, didn’t come from the established pioneers at Atari. It was made by upstart toy company Nintendo, and wasn’t even designed by a professional coder, it was created by Shigeru Miyamoto, an aspiring cartoonist. In all trades, in all workplaces and at every age, cross-training and cross-disciplinary learning is vital. The best discoveries come from those who are hungry to learn, not those who sit soaking in what they have learned.
Hunger to learn versus hunger to be learned can be seen in every technical field. If you went to college for a specialized subject you’ve seen it. There are the masters who can recite every equation, muscle group and atomic mass – the rote learners. And then there are the masters who dig into comprehension and mechanisms, the ones who can derive the equations, visualize the function of each muscle group and see how the atomic weight informs the material properties.
Expert decision-making can easily degenerate into fast-thinking repetition from muscle memory. The chess master Herbert Simon criticized the glorification of expert intuition when he said:
QUOTE “The situation has provided a cue; this cue has given the expert access to information stored in memory, and the information provides the answer. Intuition is nothing more and nothing less than recognition.” END QUOTE
Semmelweis wanted more than memorizing and reciting answers: He wanted to understand. He experimented and extrapolated and went wherever it took him. At the end of the day he was a scientist first and a doctor second, and many of his colleagues were barely doctors at all.
So what of Billy Mitchell’s grand defense? Why would an expert cheat?
Because experts are the most effective cheaters. In any competitive arena, the most successful cheats and cons are inevitably done by the ones that best know the craft. Gaming speed-runner and historian Karl Jobst put it best;
QUOTE “You might think that having the talent to achieve a world record would make someone less susceptible to cheating, but it often works against them.”
“They feel like they deserve the world record, and when the game they play doesn’t give them the luck they need, they become increasingly frustrated. On top of this, having talent and deep knowledge of a game makes you a better cheater. You know what tools to use and how to hide your edits. You understand what does or doesn’t make sense, what is or isn’t possible, and what’s believable. When top players cheat, it’s almost always detected by other top players looking much deeper than a normal spectator might.”
“The fact that better players make better cheaters means that arguments such as ‘he’s such a good player, he has no reason to cheat’ are fundamentally flawed.” END QUOTE
The incentive to prove and protect your expertise can undermine it in an instant. Billy Mitchell was routinely within spitting distance of the world record, so he employed his expertise to just take it. The fact that he could have theoretically done it for real was all the justification he needed. Semmelweis’ peers worked too hard to get where they were to let an upstart with wild ideas make them feel inferior. They were doctors – to think they were making people ill was an insult to their entire identities.
Experts are hugely important, vital even, to all of society. But their self-perception and public perception can lead them down dangerous paths: Experts are not inherently geniuses and, like any human being, they are not perfect or infallible. The best experts are constantly looking to expand their understanding, and the worst are constantly trying to prove their worth. Both kinds will make mistakes, but only the former will admit to them and learn from them. Every human being has a range of their own expertise and must govern themselves similarly: respect the limits of your own knowledge, be grateful to those who can fill in the gaps, and don’t put anyone on a pedestal, especially yourself.
Thanks for listening. As usual I’d like to remind you that all of my illustrious job titles come with the prefix “armchair;” if you’re an expert, and you’re hearing me get it wrong, I’d like to hear from you.
This episode was written, performed and edited by George Frankly with additional production support by Adam B. Levine. Our theme song comes courtesy of Jared Rubens and this episode's album art features a photograph provided by Kelly Sikkema/Unsplash, modified by Speaking of Bitcoin. Have any questions or comments? Send Adam an email at firstname.lastname@example.org