What Ethiopian runners taught me about reading scientific literature
Or, how to go from overeating Italian food to winning marathons, if you are Ethiopian
Part 2 of 2. (previous part here)
Someone once told me about their friend who’d convinced MIT to pay for them to go to Burning Man.
The first time he attended the festival, he told the school he wanted to study its unique gifting economy. But when he got back, he “realized” he’d either lost some data or really needed to collect more. This meant he had to go again.
Halfway through my PhD, I was restless; I didn’t think I could go on crazy research adventures like Burning Man Guy because I was in a biophysics program. (In retrospect that was my bad, because someone in my cohort spent a lot of her degree hiking through Chilean deserts to study sweaty shrubs.)
Instead I trained for a marathon, which I failed to run. But it’s about the journey, I guess, and I did get to listen to some really long audiobooks along the way. One of these books was the offspring of anthropology fieldwork by PhD student Michael Crawley,1 and it was about how he moved to Ethiopia to train with the best distance runners in the world.2
Question: Just how much better are they?
Answer: A lot better
Crawley has a marathon time of 2:20, or an average of 5 minutes 20 seconds per mile.
One morning in Ethiopia, on a ride to a training session, Mike strikes up a conversation with the bus conductor Tadesse.3 Tadesse had tried running for a while but decided he wasn’t good enough to take it seriously. The two share race times, which makes Mike rather existential.
This is perhaps the moment I realise the extent of the gulf between me and the other runners on the bus. Even the bus conductor is two seconds quicker than me over 10 kilometres and he ran his [personal best] at 2400 metres above sea level.
At 2400 meters, using this calculator, Tadesse might have been 21 seconds faster than Crawley for a 10k and then decided he couldn’t compete in Ethiopia.4 Meanwhile, Crawley raced for Great Britain—he still carried the gear with him and everything. This was very confusing for the Ethiopian athletes (see cover art above).
“For enjoyment only”
To pile on, Mike finds out how Ethiopians really think about domestic versus international races. The following comes from Fasil, one of Mike’s training buddies.
“When you go outside,” Fasil points out to me, “there may only be six or seven athletes from Ethiopia and a few from Kenya, and the rest of the people there will be running for enjoyment only.”
I’m not sure that this is how many serious club runners would think about it necessarily, but I let him finish.
“When you run a race in Ethiopia, though, you have to fight with hundreds of strong athletes.”
Fasil basically thinks that if everyone else was taking running seriously, then shouldn’t they be better at it?
And he might not be completely wrong. Because while Ethiopian athletes have clear genetic and environmental advantages—elevation and a strong infrastructure for professional athletics, for instance5—Mike found that they trained completely differently, too.
How to run fast, if you are in Ethiopia
Through Fasil, Mike met an athlete named Birhanu Addissie.
Birhanu wasn’t performing as well as he wanted recently. Part of the problem was that Birhanu had gained about ten pounds from eating too much pizza. Mike knew this because he watched him eat a lot of pizza; also, other runners had started making fun of him by calling him “Birhanu Pizza.”
But Birhanu wanted to win the 2015 marathon in Rome, which was in five weeks.
Step 1. Make it harder
Step 1a. Find the thinnest air
To train for Rome, Birhanu needed to find the right places to run. Mike asked to tag along on one of these training runs to a place called Arat Shi. “Arat Shi” literally means 4000 meters, and this worries him.
“Surely we’re not the equivalent of almost halfway up Everest and planning on running uphill?”6 Mike asks.
I wanted to know what it was like to train at that altitude, too—but vicariously, of course. So I found a YouTuber who did an exercise challenge at 10,631 ft above sea level. He completed the challenge, but then needed aid and cried uncontrollably.7
When Mike asked Birhanu whether Arat Shi might be too difficult, Birhanu replied,
‘Hard is good, Mike. This is marathon.’
[…]
‘And we’d better take Fasil with us,’ [Birhanu] adds after a few moments thought.
‘Why Fasil?’ I ask.
‘Because Fasil,’ he says, ‘is crazy.’
‘And crazy is good?’ I query.
‘Crazy is good.’
1b. Follow a crazy person
Fasil shows up in the book a lot. Everybody finds him a perplexing sort of guy.
This is how Fasil became a runner.
Only a year prior, Fasil was looking for a way to make some more money in Addis Ababa, the city that “cleans faces and pockets”. He could play the mesinko and thought he could make a bit extra singing songs.
Fasil went to the market to buy one, but he got distracted by a shop with running shoes.
“So I put them on and jumped up and down, and they were so bouncy I felt like I could run for ever. So I spent all the money I had on the shoes and I had to run all the way back to Kotebe, because I had nothing left for the taxi fare.”
The next day, Fasil ran into the forest with his shoes and started following the first group of people he saw. They ran so far that he had to take a bus back, but he didn’t have any money.
“Luckily I met a girl who invited me in for some milk and honey on their farm,” Fasil says, to which his friend Hailye raises an eyebrow. “She gave me the money for the bus”.
“You see,” says Hailye, who had introduced Fasil to Mike. “This is why I wanted you to meet Fasil. He’s a bit different.”
This was why Birhanu didn’t just want Fasil to join the training session at Arat Shi—he wanted him to lead it. Crazy is good, and Fasil would create the runs Birhanu needed to perform in Rome.
1c. “I’M NOT COLD, I AM A VERY DANGEROUS MAN!”
A sentence yelled during a cold shower.
One morning at 3 a.m., Mike wakes up after four hours of sleep and sees Fasil washing his face at the tap.
“It is pitch black, and my breath turns to mist in the cold air as our dog barks at me,” Mike writes in his thesis. But he’d agreed to join on this run. Fasil beams at him in running gear, telling him, “You’re no foreigner, you’re a hero.”
“Are you tired?” I ask Fasil in Amharic.
“I am not tired!” he exclaims, grinning.
They do an hour of hill sprints and return for a cold shower before going back to sleep.
What was the point? Mike wonders.
The argument for being dangerous
“Disenchantment is the distinctive injury of modernity,” wrote the sociologist Max Weber in 1946.8 A major theme of Crawley’s book is how Ethiopian runners fight this disenchantment. They talk about air being “magic” and about “mysterious incalculable forces” giving runners energy and life.
Crawley’s training back home was not as magic.
We run with heart-rate monitors and GPS watches at a carefully planned pace. We upload the GPS data about our running to apps like Strava… Sports scientists test our top athletes in order to determine their physiological parameters.
Calculated optimization takes a lot of the magic out of the sport. But if you make life especially difficult once in a while, just for fun, it can make you feel like you’re on the edge of what’s possible—like no one else is “dangerous” enough to do what you do.
Back at Arat Shi, even before the run begins, they encounter some difficulties. It had rained really hard the day before and their shoes were caked in mud.
“Fasil stops to carefully scrape it off,” writes Mike, “which seems to me a futile gesture as we are about to be running through it for two and a half hours.” And then when the run begins, Mike finds out what “crazy” really means.
“Fasil suddenly decides to loop right round a tree and go back on himself, or make a 90-degree turn by ducking under a low branch.” He leads them into a ditch they have to climb their way out of, through “bushes with inch-long thorns,” and one of them lodges itself in Mike’s head.
All in all, it’s exactly how Birhanu wanted his training run to go. But not Mike, who’s so frustrated he wants to peel off and choose his own routes. Fasil says he can’t, though.
“‘No, no,’ he tells me. ‘There are way too many hyenas around here.’”
Mike looks back at Birhanu to see how he’s doing, and Birhanu is “floating along with a serene look on his face. He is wearing two tracksuits to make things a bit harder, but there isn’t a single bead of sweat on his forehead.” He sees Mike checking in on him and smiles back in encouragement.
Two-and-a-half hours and 19 miles later, they stop to cool down. Mike writes, “The outsides of my shins are throbbing from the effort of staying upright,” but “Birhanu bounces from foot to foot like a boxer.”
“‘Tomorrow, three hours!’ he beams. ‘This air is nice!’”
2. How to run fast by making it easier
Despite the torture run, Mike noticed there were a lot of ways Ethiopians made it easier to train.
2a. Consider holding hands
After the run at Arat Shi, Mike discovers they have to walk back to where they started. He throws a tantrum.
Fasil keeps giving me concerned pats on the shoulder.
‘Dekemah zare?’ he asks me every couple of minutes. ‘Are you tired today?’
‘Yes,’ I tell him in English, ‘and I could have done without you choosing the most difficult, dangerous route possible, and without having to trek for hours to get home!’
He smiles wryly and just says, ‘Ayzoh, ayzoh,’ (keep going) and my irritation gradually lessens as we walk down the hill. I feel like a petulant child on a long car journey. I trudge slowly.
Fasil looks absolutely fresh, and walks with a spring in his step and a swing to his arm. As we walk down the hill he takes my hand affectionately. Ethiopian men often hold hands, something that initially surprised me but which I’ve got used to now.
‘Ayzoh,’ he says again.
That run was hard, but it’s over now and we can rest.
2b. Zigzags for everybody
“Why did you want to run alone?… Training alone is just for health… To be changed you have to run with others.”
For Mike, the most dramatic difference between running in Ethiopia compared to the UK was that nobody ran alone. In fact, it’s seen as “deeply antisocial and borderline suspicious.”
This feature of Ethiopian running turned out to be really useful, because just like Fasil, Mike found his training group by going into the forest and following the first group he saw. Unlike Fasil, though, he couldn’t keep up.
Before I know it, the line of runners comes up alongside me and I am grabbed by the wrist and guided back to my place at the back of the queue.
‘We run together,’ the runner closest to me says in English.
I resign myself to following the group, and the purple tight-clad pied piper at the front, as best as I can.
The strangers yell encouragement at him when he slips behind again— “Ayzoh, ayzoh!”—and finally, when the run is finished, he introduces himself to a runner named Tilahun.
Tilahun has the following questions:
Where are you from?
Why are you here?
What is your 10k time?
Why did you want to run alone?
Later, Mike’s new friends bring him to a training field. To his astonishment, he sees a few hundred runners with no separation by ability—kids were running next to 2:05 marathon champions, and he thinks he spots a teenager who’s just wandered over from a nearby farm.
The train of runners meanders around the field, the leader turning almost 180 degrees every minute or so and the rest following like a shoal of fish.
‘Why do they zigzag like that, Meseret?’ I ask.
‘I don’t know,’ he replies… ‘They learn that from each other. No-one is telling them to run like that.’
Mike’s theory is that the bright athletic jackets attract newcomers like flowers do bees, and the zigzagging lets them cut corners while the experienced runners don’t. Whatever the reason, these training sessions let everybody run on the same field. And after a hard practice, they sit on the grass and hang out, all together.
2c. “As interesting and inspiring as possible”
When I was in Cambridge, MA, I always ran the same route along the Charles River. I ran at the same time on the same sidewalk, and if I were in the mood, I might go clockwise instead of counter-clockwise.
To cope with the boredom I watched the geese that swarmed the riverbanks, taking pictures of them to post on Strava with updates about their social lives.
But in Ethiopia, coaches worked hard to make the training itself engaging. Running on asphalt was restricted; it was much more important to learn how to respond to the texture of rocks, or gravel, or overgrowth on the ground; to learn how to lean into the way the land made you move.
“The successful ones,” a coach says, “are the ones who watch with their eyes and think with their minds before they move their legs.”
Some time before Rome, Birhanu moves onto gravel road. Even the gravel is uneven and difficult—the roads are “never flat and you still have to think carefully about where you are putting your feet. It is still very hard work.” And after five weeks of 3 a.m. hills, hand-holding, and hyena bramble torture runs, Birhanu sidles over to Mike one day and says,
“I think I am winning in Rome.”
And he was very, very close to being right. Birhanu “Pizza” Addissie won silver in the 2015 Rome marathon, followed immediately by gold in Taiyuan.
A very smooth segue
Now I’m going to see whether I can learn from all of this and apply it to my own job as an academic researcher. Because running until you throw up to become a world-class athlete is kind of like sitting at a desk reading papers, maybe.
What if we did more together?
Reading papers is boring, but I realize now that I always read them alone.
I’ve never done a lit review with somebody, in real time. Wouldn’t it be nice to watch an expert look through the literature, extract what they need, and see how they decide what matters?
As a counterexample, look at software engineering: in classes or jobs, there’s a concept called “paired programming” where you share your screen with someone and they watch you code. The first time I did this for a class I hated it—my partner had worked at Google and I’m self-taught, which made me very self-conscious. But then one moment changed my mind.
It was when, during a class session, we came across a line like this:
acts, weights, noise = np.zeros(n), np.zeros((n,n)), np.zeros(n)
He clicked his tongue and said, “This is why Python is pretty.”
And I thought, huh. I guess it is.
What is the reason for being dangerous?
For every boring thing out there, there are people who have convinced themselves it’s not that boring. This is the whole premise of meditation. And to get yourself to do boring things, it helps to feel some sharp intensity of purpose behind them, to feel that the importance of your goals are exemplified by the crazy lengths you go to achieve them.
But for scientists, what is that goal? Why are we putting ourselves through the gauntlet of academic language?
For the Ethiopian runners, the goal was to win races and break records, which they did by training in thin air and on tricky ground—but that wasn’t where they stopped. To run really fast—to actually break those records—they had to get off the mountain.
What is science for?
Scientists are supposed to be learning, discovering, creating concepts that other people can leap off to create even more. There are so many possible inventions and undiscovered truths, and we’re supposed to be trying to work them out, together.
Instead, we train and train by reading these indecipherable papers; then we graduate and spawn more of our own. It’s like we’ve all decided that running doesn’t count unless it makes you cry by yourself at 10,000 feet above sea level. If that’s the sport we’re trying to win, defined by endless nonsensical obstacles, then today’s scientific writing style is perfect; it makes us feel good and powerful and dangerous to tell ourselves that only we can perform at these heights, whatever that means.
But look. Clearly that’s not the point. And obviously there’s some problem with how we’re doing scientific communication, because fact that there are anti-science movements out there reflects that we’ve done a really shit job explaining what we know, because “science” is just what we think is probably true about the world, and why we think those things. The whole point is to (1) find stuff out, and (2) make it make sense.
So shouldn’t we all just get off the mountain and write in a way people can (and want to) read?
If you want to make a scientist angry, explain something to them poorly. I know this because I have done it a lot. But it’s not just scientists; it’s that almost everybody gets frustrated at what they don’t understand. And despite my complaints here, I’m still part of the system and I’m still writing these papers. I love doing scientific research so much that I’m making it my career—but the reality is, I don’t even like reading my own work.
So why should anyone else listen to me?
Crawley, Michael. Out of Thin Air: Running Wisdom and Magic from Above the Clouds in Ethiopia. Bloomsbury Publishing, 2020.
Because Kenyan runners tend to get more press than Ethiopians, Crawley explains why he went to Ethiopia instead (it’s because they’re better).
Since 1960,
Ethiopian men have won twice the number of marathon gold medals as Kenya.
They occupy six of the top ten spots on the all-time marathon lists.
They have won five Olympic 10,000m titles since 1980 to Kenya’s one, in spite of boycotting the event twice.
Ethiopian men and women hold all four 5000m and 10,000m world records.
Names in Ethiopia go [given name] [father’s name] [grandfather’s name]. They tend to use given (first) names or else things get confusing, so Crawley uses them in his book, too.
Thanks to user bertrand russet for the correction to my original version
Some claim that East African runners benefit from coming from poor, rural backgrounds and having to run to school every day, but athletes in the book point out that you need money, time, and facilities to train properly.
One athlete in particular “is unconvinced by the idea that running ability comes from having to run to school, pointing out that [world record holder Kenenisa Bekele] lived in the town’s main square, a stone’s throw from the school”.
Mike is right; it’s actually more like 2500 meters, or over 8200 feet. But still, “it is the sort of altitude at which you get a headache as soon as you start running and at which, it turns out, I start to lose the feeling in my fingers.”
And it’s the altitude where conversations like this happen after a few laps:
‘Breathing,’ he starts to tell me, before raising a finger in matter-of-fact excuse and turning to quietly vomit on the grass, ‘really burns here. It’s not like in Addis.’
I’ve had a pet hypothesis for years that altitude makes people emotional. This was based on my own experiences (I’m skeptical other people found Marvel’s Eternals to be a deeply profound movie, unless they also watched it on a plane), based on friends’ stories, and lines from the book Far and Away by Andrew Solomon (2016).
I used to think that I was unusually reactive to the thinness of the air in a plane’s pressurized cabin. I cry on planes—at the movies I watch, the books I read, the letters and e-mails I attempt to answer… I sought research that would reveal whether more or less blood was flowing to which areas of my brain, how my pulmonary capacity was compromised by the angle of ascent.
And the research is there—studies suggest that moving to higher elevations increases depression and anxiety (2019), and that staying there makes you less likely to think other people are happy (2025).
But Solomon wrote this before these papers came out, so he had to come up with his own reason for getting sad on flights.
Now I’ve come to believe that departure simply makes me sad, whether it points to someplace I’ve always wanted to see or to the home I have missed. Though travel can intensify life, it also evokes dying. It is a detachment. I grow anxious at takeoff not because of the air pressure and not because the plane might crash, but because I feel myself dissolving.
I’m glad Solomon couldn’t see the future of research, though, because now there is this lovely essay. Plus, he was probably a little bit right, and a sudden decrease in oxygen does “evoke dying” to an unprepared brain.
Weber, Max. Essays in Sociology, pp. 129-156, New York: Oxford University Press, 1946.
This was very fun, thank you, but I don't think that the +30 s/mi figure at 8 kft altitude is the correct figure in this case. The best source I was able to find is a calculator (http://www.runworks.com/calculator.html) that estimates +3 s/mi for a 5 minute mile at 8 kft for altitude-adapted runners. The bus conductor's 10k PR likely would have been 21 s faster than Crowley's at sea level, not 3 minutes.
I think the reason that the expert cited in your source gives the time as being "up to 30s" is that he assumes readers are both slower and less altitude-adapted than Tadesse, both of which increase the time difference.
Granted, 21s is the difference between 1st place and 14th place in the last two Summer olympic 10k finals, so the broader point stands.
Dear Chenchen, can we translate part of this article into Spanish with links to you and a description of your newsletter?