The craft of economic storytelling
Episode Transcript
CARDIFF GARCIA: Hi, I'm Cardiff Garcia, and you're listening to the New Bazaar. Coming up on today's show.
TIM HARFORD: It's not enough to just explain it clearly, you have to, to do more than be clear. You have to actually awaken people's curiosity.
CG: Tim Harford on how to tell a great story.
Tim Harford is the author of a bunch of my favorite economics books. And he's also the host of two great podcasts, Cautionary Tales from Pushkin Industries, which is about what we can learn from big mistakes, and the BBC's More or Less podcast, which is about statistics. Plus, he writes columns and essays for the Financial Times so he just does a lot of stuff. And what sets him apart in all these different mediums is that he's just an exceptional storyteller. And when it comes to telling stories about the economy, stories that are meant to grab your attention, and keep you in suspense and then ultimately land on a message or a lesson that really stays with you, I'm not sure there's anyone better.
Tim's latest book, The Data Detective, is no different. There's a lot of great stories in it. But what I really loved about it is it's very much also about the craft of storytelling itself. And that's what today's conversation with Tim is about. What are the ingredients of a captivating story? Why is it that stories are so necessary for fighting back against misinformation? Why aren't the facts themselves enough? And how do you wield the power of storytelling responsibly? There is a specific reason that Tim has been thinking deeply about these questions for the last few years. And we talk about that, too. So in this chat about storytelling, you'll also hear a little bit of Tim's own story. Here it is.
CG: Tim Harford, welcome to The New Bazaar.
TH: Oh, so pleased to be on the show. I'm sorry, I don't have a Nobel Prize like some of your guests, but I'm gonna do my best.
CG: Oh, Tim, come on. I mean, surely the distinction of being the world's best economic storyteller is better than a Nobel Prize. Damn the Nobel Prize.
TH: Well, it's, it's really ... There are some good economic storytellers out there. So I'm, I'm very flattered. I'll take it. I'll take the compliment. Thank you.
CG: Yes, yes. Here's where I wanna begin. There is a very famous book called How to Lie with Statistics. It was written by a guy named Darrell Huff, and it was published in 1954. This was a book that you once loved and read and reread over and over. And then later, you came to dislike it because it is so cynical about the use of statistics. It kind of treats statistics as like weapons that are used by the bad guys. And in your own book, The Data Detective, here's what you write, "What does it say about statistics and about us that the most successful book on the subject is, from cover to cover, a warning about misinformation." So let's start there. What does it say about us?
TH: Well, it says that we've become extremely cynical about statistics. And Darrell Huff wasn't the first person to, to play into that cynicism. There's a famous old quote, sometimes attributed to Mark Twain, sometimes attributed to Disraeli, I don't think it was either of them. “There are three kinds of lies: lies, damned lies and statistics.” I mean, this is, that's a 19th century quote. So people are suspicious of statistics. They think that statistics can be used to lie to them. And of course they can. But i- it would be weird if I were to publish a book about language that was titled How to Lie with Words and I just said to everyone, "Oh, you need to be careful of words 'cause people, people lie with words."
And everybody just said, "Oh, yeah. Words." When we're not having anything to do with words in the future, because people sometimes tell lies, it would be absurd. Because of course, words are useful for so much more now than lying, even though they are, of course, u- used for lying. And it's the same with statistics. Statistics are absolutely vital, both for communicating facts about the world and for discovering facts about the world.
And I think part of the problem is that we, we feel maybe wrongly that we have the capacity to tell the difference between truth and lies when they come in the form of words. But when they come in the form of statistics, a lot of us feel helpless. And we just go into this reflexive defensive crouch and we just say, "Oh, I guess I'll just disbelieve any statistical claim." And that feels kind of smart and savvy, but actually, you're turning your back on some really important stuff if you do that.
CG: Yeah. And by the way, the author of that book, Darrell Huff, as you wrote in your own book, ended up being like the expert witness for the tobacco industry some years later after he published the book because he was helping them to essentially sow a lot of doubt about some very well-arrived at statistics that linked cigarette smoking to cancer and to other bad things. And I kinda couldn't believe it when I saw that. I mean, here's this guy who had written all about how to lie with statistics, who is, in a sense, helping to lie with statistics. So he wasn't using it as a warning, uh, to others to, to how to you know, how to defend yourself against the bad guys who are misusing statistics. He became the bad guy.
TH: Yeah, it's very sad. And, uh, and it fits very naturally with the, the strategy that the tobacco industry were using, which is rather than try to persuade people that cigarettes were safe, which is a very difficult thing to do, you just persuade them that there are reasons to doubt the evidence that cigarettes are dangerous. That they realized people smoking cigarettes don't want to believe they're killing themselves. It's kind of hard to quit cigarettes. You'd rather believe they're probably okay. And so cigarette smokers are going to easily t- eage- eagerly, rather, grab on to any evidence they can to doubt the scientific consensus that's emerging.
And, and, and so the tobacco lobby would say, "Oh, well, we need more research," which, of course, is always a good idea. They would say, "Oh, it's complicated," which, of course, it is complicated. They'd say, "Scientists disagree," and of course, scientists always disagree. Now, all those things are true. But when you add them all up, you get this message that is profoundly untrue, which is nobody knows anything, don't pay any attention to the evidence, don't listen to the experts, don't listen to the scientists, just keep smoking.
Darrell Huff's cynical view of statistics was that they, they were often used to mislead, they just fitted perfectly. In fact, he even worked on a sequel to How to Lie with Statistics, which I think, fortunately, for everybody concerned was never published, called How to Lie with Smoking Statistics. And the historians have put that together and, and, and a- assembled what a draft of it looked like. And it's up on the web. If, if you wanna have a look at you know, that half attempt to basically do the same sort of treatment for the evidence that epidemiologists were coming up with. And, and, you know, it cuts down on that, you know, o- on that evidence. He was very good at it.
CG: And the approach that was taken by the tobacco industry ended up becoming kind of instructive for politicians and for other companies, because it was this widely studied case where sowing doubt about even really good statistics, really good data seems to work, which is a kind of depressing thought but there it is. That is, that is, in fact, the history of it. I- it ended up being a case study for how, in fact, to lie with smoking statistics. And, and there were some lessons in that. And I, I guess I understood your book as trying to kind of overturn that thinking that actually statistics can be used to understand the world clearly. And then, in fact, it's not just that they can be used to understand the world, the world clearly, it's that very little else is as good as statistics at helping us to capture real truths.
TH: Yeah. For, for me, statistics are like a kind of scientific instrument, like a telescope for an astronomer or an X-ray machine, or, you know, a particle accelerator. They're you know, they are scientific instrument that shows us some, some things that we can't see in any other way. And when you think about the modern world, so much of our understanding of the modern world, including economics, which is my particular interest, but lots of other things as well. Um, you know, do the vaccines work? W- where is COVID spreading, all of this sort of stuff. Y- y- you need statistics in order to show you these things. You can't see them through personal experience. There's nothing else that you can use to see these invisible, uh, truths. You gather them through statistics.
And I see Huff's projects, and the tobacco industry's project and then, then later it's a, it's a climate change denial project, and then later a COVID denial project, it's a project of trying to discredit this tool that we have and say, "You can't trust it. It's probably just a trick. Leave it alone. Don't touch it. Um, just rely on your gut, and that's better than looking at the statistics."
CG: There is a really terrific and deeply researched essay on your website that's titled “The Problem with Facts”, and you wrote it in 2017. I think it first appeared in the Financial Times. And just to summarize it very briefly, it, it was a long essay, the problem with facts was that essentially presenting them simply as facts tends not to be persuasive for overturning falsehoods. In other words, falsehoods that people believe and it might be catchy or easy to remember, they're very hard to overturn just by telling somebody what the truth is and pointing to data or to statistics.
Because in many cases, the truth, when it contradicts something that you believe deeply, it can feel kind of threatening. And so pe- people end up finding a way to resist the facts. I, I reread that essay recently, in light of your book, The Data Detective, and it seems like that essay kind of contained like an early iteration and early version of your book's motivating impulse. Am I right to draw that connection?
TH: So you're absolutely right. Uh, I, I wasn't planning to write a book before I wrote that piece. But having written the piece, I started to realize that there was, there was something here. I, one of the things that I, that I do, Cardiff, as you, you may know, is present a BBC radio show called More or Less. And it's a, a radio show about statistics and how we use statistics to understand the world. And because I've been doing that for 15 years or so, people have said, "Oh, why don't you write a book that's kind of like More or Less?" And for a long time, I, I didn't think I wanted to do that. I didn't think I had anything to say that hadn't already been said dozens of times brilliantly by other people. Like a guide to statistics, a lot of people have written lots of good guides to statistics.
But what I realized having written the essay you refer to, “The Problem with Facts” is that there are a couple of things that were, that were missing from most guides to statistics. Uh, and the first was this positive case for statistics, reminding people what a vital tool they are. Because so often, even the statisticians who love statistics, they get drawn into our, our ... I'll give you some examples of statistics being misused 'cause that's kind of fun and engaging and that's, it seems to be the most enjoyable way to communicate about the subject. But it gives you this false impression that that's all you ever get out of statistics, is misinformation. So I wanted to make this positive case to remind people how important statistics could be, how useful they could be.
The second thing, which also ties into “The Problem with Facts” essay, was some psychological realism about how people filter and process facts or other claims that are put in front of them. Because I came to realize in 2016, uh, and a lot of us came to the same realization at about the same time for much the same reasons, I came to realize that you can tell people the, the facts, it doesn't mean they listen to you. It doesn't mean they're interested, it doesn't mean they'll believe you. Uh, and there are various reasons why that is true. And if I wasn't willing to grapple with those reasons, then what was the point of doing any of this? What's the point in telling people about correlation and causation if they just don't care or, or will ignore everything you say?
CG: Yeah, I'd love to read a brief passage from that essay, which I think touches upon a lot of what you just, uh, what you just noted. Here's what you write, "What we need is a Carl Sagan or David Attenborough of social science. Somebody who can create a sense of wonder and fascination. Not just at the structure of the solar system or struggles of life in a tropical rainforest, but at the workings of our own civilization: health, migration, finance, education, diplomacy."
“One candidate would have been Swedish doctor and statistician, Hans Rosling, who died in February. He characterized his task as telling people the facts to describe the world. But the facts need a champion. Facts rarely stand up for themselves. They need someone to make us care about them, to make us curious. That's what Rosling did." And I would argue that is what Tim Harford does. Is it not?
TH: Uh, well, I'm very flattered by the comparison. It's what I try to do. And, uh, I've started to get more conscious about this, that i- if you want to explore, for example, uh, an idea in the economy and how the economy works, it's not enough to just explain it clearly, you have to be more than be clear. You have to actually awaken people's curiosity. You have to get them interested and, and get them realizing there's a gap in their knowledge that they perhaps, you know, never even realized. "Oh, hang on. I've never even thought about, uh, you know, how, how does stuff actually come from China and how does the shipping industry work," for example.
And this is the sort of thing that, uh, Cardiff, your former colleagues on Planet Money do so well. Just get people interested in these stories, get people curious. It's, I'm, I'm not the only person doing this. But I think it's a very powerful way to communicate. Uh, and the storytelling is a great fun. It's a great challenge as a journalist to, to find that story that also communicates the essence of the idea as well. But, yeah, sure, that's what I'm trying to do.
And, and I think i- it is an effective way of, of communicating any complex information. I think Orson Welles said something along the lines, I quote him in the book, he said, um, "Don't worry about an audience not understanding something. Audiences can understand anything, the problem is to interest them. But once you interest them, they can understand anything in the world.” So that's where I started to turn my attention.
CG: Yeah, I love that quote from Orson Welles. And I wanna turn now to the question of how to get people interested and how to induce that curiosity that you just described, and which can be maybe the only really potent thing for getting people to open their minds and to see the truth and to see facts, uh, as what they are. Not as threats to their deeply held beliefs, but as things that they should be, you know, willing to absorb, willing to accept. Uh, here's a quote from The Data Detective that comes at the very beginning of the book. You write, "The book is full of fun stories. How Florence Nightingale started a public health revolution with a pie chart, how the entire Dutch art world was fooled by their own wishful thinking, how a stripper and a congressman changed the face of US statistics, how the world's two greatest economists tried and failed to see into the future."
It seems like stories for you really are a way to induce that curiosity because stories provide that kind of enchantment that can lead to curiosity, that sense of kind of wonder and fascination that you write about. I- is that why it was important to set up The Data Detective this way, to write it this way? As a book that's full of stories and not just as a kind of simple explainers, statistical concepts?
TH: Yeah, I mean, I just love researching stories and I love writing stories and I love reading stories. Uh, and, and so it, it didn't seem to make any sense to me to do it any other way. Uh, it's just how I like to write. Uh, and I have, I have liked to write for quite a long time. Um, but, uh, yeah, I, having found some good stories, I certainly wanted to begin the book by telling people, "Hey, no, no, it's gonna be really good. There'll be some stories or comments.” So yeah, I guess that's what I was doing there.
CG: But we should note that you tell true stories, Tim. You're not a, a fiction novelist. And I guess I'm wondering to what extent you feel the constraints of number one, having to tell the truth, and number two, having to make the truth itself interesting, um, without sort of, without oversimplifying things to the point where you might be saying things that are strictly true, but can be a little bit misleading. So like that need to capture nuance in addition to the truth.
It sounds like you get a lot of joy from telling stories. But it is worth noting that you have to do this within the confines of the truth. And especially when you're combating mistruths, you start with a disadvantage. And I guess, I'm, I'm wondering to what extent that's in the back of your head when you sit down to write a story or tell a story, you know, on your podcast.
TH: I don't regard it as being a disadvantage. Uh, the, the facts are already out there. Uh, normally, the stories that I tell have been very well researched already by historians. So for example, Florence Nightingale, there are so many great biographies of Florence Nightingale. So it's easy to pick up half a dozen good books on Florence Nightingale and to, to look into every angle of her story. And then it's just a case of picking the stuff that's important to the particular story that I'm trying to tell, which is about her use of data visualization and, and how data vi- visualization was, was important in, uh, winning a, you know, a big sort of political battle in the UK at the time.
But the, I mean, the details are delightful. Her quotations, the way she used language i- is delightful. So I don't see that as, as a disadvantage. I see that as a resource. That's something you could, that you can build on. I mean, there was, with Fl- Florence Nightingale, there was one thing, one annoying little bit of truth that I stumbled into, which is there's a very famous quote. Florence Nightingale's collaborator, the great statistician, uh, William Farr, saying to Nightingale, "You complain that your report would be dry. The drier, the better. Statistics should be the driest of all reading."
And this is in several biographies. It was in a piece published by, uh, The Economist. Um, and in fact, I put it in one of my own pieces, because it's this great idea of, of this, you know, this dusty, old Victorian statistician. You can just imagine him with this big white beard, looking a bit like Charles Darwin. And he's telling the fiery young campaigner, Florence Nightingale, "Make your statistics drier." And she ignores him and she produces this amazing data visualization that changes the world. That's not true. None of that is true. And I realized 'cause I stumbled ...
CG: She wrote it.
TH: She wrote it to him.
CG: Yeah.
TH: She told him statistics should be the driest of all these reading. And, and I stumbled upon this one. I, having seen this quote a dozen times, half a dozen times, I was reading a biography of William Farr. And in the biography of William Farr, it's the other way around, she writes to him. So then I just had to get to the bottom of this and I wrote to the editor of Florence Nightingale's, uh, papers, and I, you know, consulted the, the British Library. And, and yeah, and I figured out no, she, she said that to him.
So then it's much more complicated, like, "How, how is it that she produces this beautiful data visualization but she's also saying to William Farr make it super dry? What, what was going on? Is that a contradiction? Did she change her mind?" I mean, I guess I could have just ignored it. But I, I was kind of pleased with myself that, that I found this thing that all these biographers of Nightingale haven't found.
CG: You found it!
TH: So i- in the end, I think I, I think I figured it out what she's really saying there and they're very much in agreement on a lot of things, what she's really saying is, "If you're participating in a political battle and you're using statistics, you've got to make sure that the foundations are impregnable. You will be criticized. People will come for you. And you just got to make sure everything is rock solid. And then once you've done that, then on top of that, you can editorialize. But first, get the statistics right. And then you can start doing the, the messaging." That's what I think she was saying. Maybe I'm wrong.
CG: But that, that lesson is important for other reasons, too, which is that ... Especially as a writer or as a journalist, people who are tasked with telling true stories, there's always a kind of temptation to gravitate towards the thing that is immediately fascinating or interesting. And there's a temptation to oversimplify because, of course, caveats and nuances can kind of get in the way of a great yarn sometimes. And it can be really tough. And I, I remember on The Indicator, which was a podcast that I, I hosted before the New Bazaar, co-hosted with Stacey Vanek Smith, a lot of the stories that we tell there are narrative stories, you know. They, it was storytelling and not just like questions and answers and things like that.
And for those narrative stories, you know, you, you'd hear something or there'd be like this nice, clean co- you know, counterintuitive thing. You think, "God, what a great story." And then you discover something that complicates the story. And that's hard to insert without losing like the narrative thrust, like the thing that made it so fascinating. But Stacy and I really cared about getting things right. And so we would do our damnedest to make sure that all of the nuance and complexity was captured there.
But it was really hard. It was very seductive to just go with the thing that makes it so captivating and you have to resist that. But it can be hard. So I guess I'm kind of curious to know what your approach is to reconciling those things. You know, the, the thing makes the story so fascinating versus the complicating nuance, the complicating detail that is really important but you got to include it in there even if you lose some of that narrative momentum.
TH: Yeah. It's, it's always a judgment call because, of course, you can't include everything. And, you know, is this an essential detail? Or is it just something that's gonna trip everybody up? And that also depends on the medium. So some of the stories, I've had the joy of telling twice. So once in The Data Detective and once in my podcast, Cautionary Tales, there's an episode of Cautionary Tales devoted to Florence Nightingale. The story about Irving Fisher and John Maynard Keynes, there's a Cautionary Tales about them. And the, the challenge, I think, of communicating an audio is you have to keep it even simpler.
Because people can't just go, "Oh, hang on, I missed that. Let me just reread that sentence." I mean, people can rewind on podcast, but they, they don't. So there's always this trade off. Um, for me, I often find that on the last pass through, I'm stripping away details, particularly in, in audio. Because there's a certain simplicity you need. But it's always important to make sure that the details are not the thing that ...
If you told somebody this detail, would it totally change their view of the story? Would they go, "Well, hang on. This completely contradicts everything you've just said"? Or is it just like, "Oh, it's interesting. It would have been nice to hear that detail but I guess you can't say everything"? That's the thing. You're fundamentally telling the truth, you're fundamentally giving the right impression, you can't leave stuff out that, that really makes a difference. Um, but at the same time, you, you have to leave all kinds of stuff out and you have to make choices.
CG: Yeah, I love this point also about the difference between telling a story in text and telling a story in audio. Because you're right, in audio, you know, telling a story in chronological order is a lot more compelling because people find it very difficult to sort of bounce around, in their heads, bouncing around back and forth through time. Whereas that is something that you can kind of play around with in text. And also in audio, the big difference I have found from telling a story in text is that once you lose your audience, it's over.
Once you start telling a story on a podcast or on the radio, and somebody loses the thread, that's it. They're not coming back. Whereas for something that's in front of you, you know, you can look up, you can think about something and you can go back to the text, you can look up again, if you, if you sort of tuned out for a few paragraphs while you were reading, it's very easy to go back. In audio, storytelling is very tricky. Because once the listener is gone, you're not getting them back. And so you have to be so careful in how you design the story itself so that you never lose them. And playing around with bouncing back and forth through time, it can be done but it's tricky. It has to be done so carefully.
Whereas in text, you can kind of play around. You can take like these little discursive moments, you can go off on a tangent and come back. On audio, you can't. And it actually is something that I think i- is quite underappreciated. And I separate that, by the way, from something like what you and I are doing now. We're having a discussion, we're having a conversation. I hope people don't tune out ever but if they do for, you know, 10 minutes or something 'cause they're washing the dishes and they come back, they can sort of pick up the thread of what we're talking about, because it's just a chat. But when you tell a story in audio, it's very tough to hang on to the audience.
TH: Yeah. You're, you're building more and more and more information and, and references don't necessarily make sense anymore if they've missed something. I mean, it is a great challenge, though. And, and you can cheat a little bit in audio. Of course, there's music, you can get actors. I mean, I ... Helena Bonham Carter played Florence Nightingale in Cautionary Tales. I mean, that, that's, that's like, it really feels like you call them a cavalry. And at that point ...
CG: That's awesome.
... it's like, "Oh, I just have Helena Bonham Carter." I mean, that's amazing. Uh, this is, by the way, this is not something I thought I would be able to say about my life, that at some stage I, I wrote script for Helena Bonham Carter.
CG: Yeah.
TH: And, but it, it is interesting that, that I've been able to go back and, and rework some of these stories in a different form. And very often, it's surprising. You read a book chapter that's 8,000 words and you think, "How am I going to get this into a podcast script that'll be 5,000 words?" And you start reading and you go, "Well, first of all, like that ... It's easy. There's like 3,000 words that I don't know why I originally put this in the book." Um, and of course, in the book, they make, they make sense. But for a podcast, I'll get to that, can't get rid of that.
But it's also surprising how often you see o- on a second viewing that you, as journalists would say you buried the lead. Like, why on earth did I think it was good idea to start there, when, obviously, obviously, the crucial point in the story, the moment you wanna start is here instead? Uh, i- it's surprising how often you realize there's hey, there's a different way to do this. Not necessarily a better way, just a different way. And you're finding that and that's part of the joy of being a writer.
CG: Yeah, that is a really fun moment of realization when you've had the space to tinker with something. Maybe you go away from it then you come back and, uh, it just sort of it hits you all of a sudden, like, "Oh, my God, look at this fun thing I can try with this story that will make it so much more compelling so that it starts quickly and then doesn't slow down," essentially. That, I found that to be very helpful especially on audio. Um, to really grab somebody early and then don't let go. You know what I mean?
Sort of, you know, don't, don't sort of like go throughout the whole story because once you've got somebody, it can be a really special thing I have found.
TH: No, absolutely. Absolutely. And because we're interested in ideas, I am, you are, listeners to the New Bazaar are, of course, here for the, here for the ideas as well as the as well as the storytelling, um, you know, you're trying to get ideas across as well. And then the question is, "Well, what's the optimal mix?" And for something like The Data Detective, it's probably 30% story and 70% ideas. And I bet this story is there to kind of keep people reading and it's kind of fun. But actually, we, I'm, I'm trying to make a particular case here. Whereas with Cautionary Tales, it's probably 70% story and 30% ideas. That you, you know, you wanna drop an idea in there because that's part what the podcast is about, but it's basically a story. Uh, and that's fine, that's fine.
CG: What to you are the ingredients of an interesting story?
TH: Well, I mean, the more I think about it, the, the more influenced I am by just the classic view of this, that, which is that there's, there's a protagonist, right? Somebody is doing something or is trying to do something and is overcoming obstacles. And maybe they're a hero, maybe they're a villain. But there's a protagonist at the center of the story. There's a twist like it shouldn't be obvious from the start how the story is gonna end. Or even if it is obvious how the story's gonna end, there needs to be a question as to w- why is it gonna end that way? And often with Cautionary Tales, you know it's gonna end a disaster, but you don't know why. Why, why is stuff gonna go wrong?
So there's that, there's that sense of, of identification with a person. Yeah, you're viewing the situation to a particular person's eyes that brings you along. And then there's the sense of you don't know exactly how it's gonna work out. There'll be some twists, there'll be, there'll be triumphs, they'll be disasters, but you don't know how it's gonna play out. There's a, and then therefore, it need, it needs to be this narrative arc. So that's your basic ingredients for a story. There's nothing mysterious about that. Of course, the fun stuff is when you start messing around with, with those or, or breaking the, the, the standard form because you have multiple protagonists or, or it's, it's perfectly clear what's gonna happen right from the start, but you're doing something else. So it's fun to play around with.
But I think a lot of people when they talk about stories ... I mean, not, a screenwriter would never talk like this or a novelist would never talk like this. But I think a lot of people in nonfiction, a lot of people doing business writing, a lot of people doing journalism, when they say story, what they mean is anecdote or example. Like, "I need to give you an example of this thing that I'm talking about."
And that's fine. I mean, it's gre- it's great to have an example. It's great to have an anecdote. That's not the same thing as a story. That's more like, "I'm just gonna give you something concrete to catch your interest or, so that you understand what it is that I'm saying." And that's totally fine. But a story is keeping your interest sustained for page after page after page, minutes after minutes after minutes of audio. And that, that's a different thing.
CG: Yeah, that point about not knowing where it's gonna end up, the suspense is to me the key distinction between what makes an idea interesting and what makes a story interesting, what makes them different. Because I think that ideas versus stories, well, they, they actually share some things in common. And I, I want to just share something with you I came across recently, Tim, um, and get your thoughts on it. I was doing a little bit of reading in advance of our chat today. And I came across something written by the psychologist Adam Grant. He was writing something about what makes your colleague Malcolm Gladwell, your colleague at, uh, Pushkin, an interesting writer, an interesting teller of stories.
And he came across, uh, a sociologist named Murray Davis. And he cited him and what Davis had said many decades ago was that an idea is interesting when it is surprising. It helps if the idea is somehow applicable so that you can kind of, you know, you can associate with the, the protagonist in the story. And then, and this to me was the most interesting part of this, an idea is found to be interesting when it doesn't challenge people's strongly held assumptions. In other words, when it is not threatening as the truth can, can be very often when people have come to believe falsehoods.
I want to just quote briefly something from Davis. Again, this, this came through Adam Grant. But here's what the sociologist Murray Davis wrote, "Those who attempt to deny the strongly held assumptions of their audience will have their very sanity called into question. They will be accused of being lunatics. If scientists, they will be called crackpots. If the difference between the inspired and the insane is only in the degree of tenacity of the particular audience assumptions they choose to attack, it is perhaps for this reason that genius has always been considered close to madness."
And I, I think that could apply to stories as well. And it seems like this is what makes it hard to tell a story where you're trying to convey a truth that people might be resistant to. You have to make it interesting to induce that necessary curiosity. Uh, you have to make it interesting in a way that doesn't make it seem from the very beginning like you are gonna challenge somebody's strongly held assumption even though that is maybe what you intend to do. And so in a way, even truthful storytelling can be like this kind of act of subterfuge. It can itself be, I don't wanna say misleading in its intent, but its intentions can't be too close to the surface. What do you think?
TH: I, I, I think that's absolutely right. I mean, the thing about a story is you, you know that you're, you're going to be, uh, surprised at some stage by the story. Otherwise, it's not a very good story, is it? If it's, if, if, if it's completely predictable from start to finish, then what kind of a story is it? I mean, I know a lot of Hollywood blockbusters feel predictable in that the, you know the emotional arc, you know the kind of thing that's gonna happen. But even then, it's, if it's any good as a movie, you can't see quite how it's going to happen or quite where it's coming from. There'll be some stuff, there'll, there'll be a betrayal, there'll be a twist. Um, it's interesting that, that you, you bring up Malcolm Gladwell. I mean, I think he's very, very good at this. And, uh, he has quite a few critics. I think, in many cases, uh, that's just jealousy, 'cause he's so successful.
But I, I had an, uh, an intriguing encounter with Malcolm about nearly a decade ago, before we were colleagues, before he set up Pushkin and I started doing podcasts for Pushkin. Uh, I was speaking at a conference and he was also speaking, he was the keynote. So I got to give my speech and then sit down in the front row and just watch him for an hour give his talk. And then the next day, in a different country, this is how the cosmocrats live, I was, I was hosting a BBC radio program, uh, on the south bank in London, in the Royal Festival Hall. Uh, I was presenting the program, I was giving a talk and Malcolm was our special guest. There were a couple of other speakers as well. Gillian Tett was one of them, she's absolutely terrific.
So I got to, to do the same thing. Give my speech, sit down in the front row and watch Malcolm give his speech. This time, his speech was 15 minutes long, not an hour and it was a different subject. But very interesting for me to see him, to see him do what he does with two different topics at two different lengths and to spot some of the common patterns. Uh, and he's, he, he makes it look very easy. And I can promise you, it's not. It's not easy to do what he does. But there are a couple of things were ... So one thing he did with both talks was he set them in a very familiar context but about unfamiliar events. So one was set in the US Civil War, one was set in the Vietnam War.
Okay, so you don't need to explain to me why the Civil War matters, you don't need to explain to me why the Vietnam War matters. Okay, I'm, I'm with you. There's, and so there's no need for preamble. I'm kind of already there. Then he's talking about these slightly obscure characters and these obscure events that we haven't heard about. We're surprised, we're interested, we wanna know more. The fundamental point that he was making in each case was not so hard to grasp. So in one case, he's basically saying, "Don't be overconfident. If you're overconfident, you're gonna you know, you'll go, you'll make mistakes, you'll get caught out."
In another of his speeches, he was talking about the importance of listening really carefully and how it's easy to think you're listening but you're not listening because you're not really taking on board what's being said. So just observing certain elements of how he communicated and what he, what he obviously thought was appropriate or what he thought was probably too much of a stretch, the kind of stories he chose and the kind of stories he steered away from, I learned a lot. It's, uh, an interesting thing is that someone like Malcolm, it's not hard to see him at work. That you just download his podcast, just listen to him or watch the TED Talk or whatever. I mean, it's not a secret. Just watch it. You watch, watch the results.
But so many people don't really do that. They don't pay attention to what's going on right in front of them. Because obviously, some of that method is hidden. The research, and where he finds his stories, who he talks to, you don't see that. But you see an awful lot. It's right there. It's not like a magician's trick where everything's concealed. It's there. So maybe the rest of us should be paying more attention to the people who do this kind of thing really well. By the way, Adam Grant, he's another person who does this incredibly well.
CG: Yeah. By the way, the, the title of Malcolm Gladwell's podcast at Pushkin is Revisionist History. And that word revisionist already tells you that you're gonna hear somebody telling you something that is surprising. It's embedded in the title itself, which is a very clever, uh, way to bring people in. Um, on the point of, of a finding a way to challenge people's strongly held assumptions when people resist them so fiercely, do you think about this in your storytelling? Or are you hoping that that particular thing you're trying to achieve is sufficiently well hidden?
When it is there, uh, that people just kind of won't notice that you're here to tell a story, you hope that people will be both entertained and that they'll come away with something informative but when is the story that might be conveying a lesson that people, a lot of people, at least in the audience could potentially be resistant to? I- is that something that crosses your mind as you're crafting the story?
TH: Well, I mean, I see my role as to entertain and to inform people, not to persuade them. So I'm not here to change your mind about anything. But maybe I'm here to tell you something you didn't know, tell you a story you hadn't heard, and give you some facts that you hadn't encountered. And if that changes how you think, that changes how you think. But I don't see it as my duty to change how you think, because that turns out to be a really hard problem. Um, and, um, so, uh, I mean, I don't write polemics very often in my columns, in my podcasts, in my books. Occasionally, maybe I'm trying to change people's minds a little bit. But I don't see that as, as my primary goal.
And maybe, maybe I just don't see that as that interesting and I also don't see it as, as terribly easy. I was quite struck when I was giving talks about The Data Detective, lots of people would ask questions about, um, "Oh, I know this, this guy and he's a total idiot who believes the stupid stuff. And how can I stop him believing the stupid stuff and get him to think straight?" To which the basic answer I have is, "Maybe just start by getting yourself to think straight." And that's actually hard enough. If you can get your own thinking straight, you're doing really well, you're ahead of most of us.
CG: Yeah.
TH: Do that and, and don't worry about other people just yet. I mean, I have some suggestions. But really, we, we're doing well enough if we can, uh, if we can think clearly ourselves without taking on our shoulders the burden of correcting the, uh, incorrect thoughts of the entire rest of the world.
CG: Yeah.
TH: So, uh, that's not what I do, generally.
CG: Yeah. And here's what I, here's what I say to that, I think there's the kind of persuasion that we're all conscious of. And then there's the kind of persuasion that we may not even be aware that we're trying to achieve here. And I think in The Data Detective, I mean, clearly, you're trying to persuade people that statistics are wonderful, that these are tools that we should all revere and that we should use responsibly. And somebody going into your book thinking in the sort of vein of, of Darrell Huff, "Well, statistics are, you know, things that unscrupulous people wield to try to convince me of something that's ridiculous."
Or, "It's all just, you know, a back and forth persuasion game or something like that." You are, you are trying to elevate the nobility of statistics in people's minds. And successfully, I would add. It's a great book, it's one that I recommend. But I, I think clearly, in this case, to, to somebody who is, you know, maybe very skeptical or doubtful of statistics, it seems like this book is an attempt to persuade them otherwise, right? Even if it's not, even if, if, you know, as you said, you write it to entertain and to inform, I mean, there is an underlying, and I would say, very passionate argument in the book.
TH: Yeah, you, you may be right. Thanks for the kind words, by the way. You, I, um, I was reading another book published not long ago, uh, Julia Galef's book, The Scout Mindset. It's great fun, really interesting. But I was quite struck by the, the fact that Galef is trying to set out a very clear argument in defense of thinking rationally. And, uh, it never occurred to me to do that. I just thought, "Well, I just assume that most of the people who buy my books will kind of already want to think rationally so I'm just gonna give them some tools that will help." But Galef is like "No, no, I don't think we can take for granted the people even want to think clearly." So I'm first going to make the case that they should and, uh, then I'll help people understand why. So, yeah, you could go further down that line, for sure. And she does. And it's, it's a book I recommend.
CG: What is the best advice you've ever come across for telling stories?
TH: Uh, I, I would actually come back to, to those things that I drew from watching Malcolm at work nearly, nearly 10 years ago. That the, the stories he told were surprising. There were stories that people hadn't heard. But they were essentially trying to make fairly straightforward points and they were in context that didn't need to be explained. So, so he sort of, he had chosen his battleground. Like, "Okay, I understand why the US Civil War matters so I'm with you. But now bring me on this journey, tell me the story I haven't heard." So there's some, something there about choosing w- where to present people with complicated or unfamiliar information and, uh, and where you don't have to do that.
CG: Yeah. One final question about the craft of storytelling. There is a kind of epilogue in The Data Detective about methods that can induce curiosity in folks. And you write about research that I, I had not seen before about a group of social scientists who had brought in some people to just simply ask them how well they understand a series of things. So they might say, "Hey, how well do you understand a flushing toilet?" And the people that they were asking would respond usually with a lot of confidence, like, "Oh, yeah, no, I know how it, I know how it works. You know, on a scale of one to seven, where seven is perfect knowledge, you know, I, it's about a five or six I know how it works."
And then the scientist would say, "Okay, write it down. You know, actually try to deliberately express how well you understand a flushing toilet." Or whatever it is that they're asking. And the people would struggle a little bit. And it turns out that that feeling of struggle made them more receptive, made them more open-minded. What lessons should we draw from this research about the importance of inducing curiosity? And perhaps more crucially, how to actually do it so that people can not just be open-minded and, and can be, you know, open to absorbing truths that contradict their beliefs but also so that they just get along better with each other, you know?
TH: Yeah, I absolutely loved this research. So this is, it's partly about intellectual humility, but it's, it's partly also about how interesting the details can be. And it's also that people can correct themselves if you give them a chance. So when you ask people to explain how a zip fastener works, and they start to realize that they had thought they knew and they don't actually. They don't really know. Um, that does induce humility. They, they roll back. They don't sort of bluster. They're like, "Oh, no, I wa- I was wrong. I thought, I thought I got it. But actually, I, I realized that actually, my understanding was extremely vague."
Um, it turns out that the same basic method works with questions of, of politics and policy. So, you know, you could say, for example, uh, "Oh, we, we, um, should ... We should introduce say, uh, a, a cap and trade scheme for carbon dioxide. That's the best way to deal with climate change." Uh, fine. Okay. Um, so what, what exactly is a cap and trade scheme? Uh, just explain to me how it works? How would you ... What's actually being traded? Is it like, is it national? Or is it international? Or how do you measure this? Who, who enforces it?
Um, and, and these are real questions. These are not kind of gotcha questions. "Just tell me how exactly you think it works." Or, you know, maybe it's unilateral sanctions on Iran or, or a universal basic income or whatever it is. And again, people realize they don't understand the policy as well as they thought they did. And that makes them moderate their views. And they start to think, "Ah, uh, I was willing to die in a ditch for this policy, because it's like this the kind of policy that my team are in favor of." Now you go, "Oh, well, now I realize I don't even understand what the policy is. Maybe it's a bit daft to be so pugnacious about it. Maybe I need to treat people who disagree with me with a little bit more respect."
So this whole process of simultaneously awakening curiosity, but also awakening humility, maybe prompting a little bit of respect for others. Um, there's a similar effect, by the way, if you ask people to participate in forecasting tournament, so Phil Tetlock and his colleagues, Tetlock is famous for his work on, uh, so called superforecasters. You ask people to try and make forecasts, you just come back and check how well the forecast went. Same thing, people start to realize they're getting a lot of forecasts wrong, they're making a lot of mistakes. They start to realize the world operates more in shades of gray. They start to realize their own judgment isn't infallible.
And all of these things tend to induce a certain amount of moderation and a certain amount of civility. So, you know, I, I, I love this. And of course, the, the process of, of writing that we've been discussing is full of this. 'Cause you're constantly coming across stories that you think, "Oh, this sounds interesting. I'll, I'll write about this. And then as you start to research, you realize, "I had no idea. This is much more complex and interesting than I thought. Um, lots of my preconceptions are shattered on the way through." So you're going through this process yourself as a journalist, as a writer.
Um, but that is the, by the way, the, the one piece of advice I do tend to give people when they say, "How do I change someone's mind?" I say, "Well, just ask them to explain the thing that they're arguing in favor of. Just ask them to just explain how it works and listen, and you'll find one or two things. Either that they, they start to realize they don't really know what they're talking about and that's an education for them. Or they do know what they're talking about and you have to listen to it and that's an education for you. But either way, someone's gonna get smarter. And while it's going on, you're asking real questions and you're getting real answers. And that's a more respectful process than the more traditional process of just getting into each other. So those sorts of questions can be helpful.
CG: That moment when you realize that you don't actually know what you're doing just yet, is at once invigorating, because you realize that the story you thought you were gonna tell is quite different from the real story. And the real story turns out to be more surprising and fascinating than you had intended or than, uh, than you'd believed. And it's terrifying
TH: Yeah. But, you know, that's okay. It's, it's terrifying if you, if you just publish and then you're starting to realize you don't know what you're, you're talking about. But if you're still in the edits, then that's great. I mean, you, you see, you're asking these questions of yourself. And then when you're working with is these are all team efforts, you're working with editors. And your editors will go, "Oh, how does this thing work?" And you're like, "Oh, I, I don't, I don't have a clue. That's a good question."
CG: Yeah. Absolutely. Uh, and, Tim, before I let you go, uh, I am gonna start bringing back a little, uh, segment that I used to do in an earlier podcast called Alpha Chat, which is to ask our guests for a long form recommendation for our audience. This can be a podcast episode, this can be a book, it can be a magazine article. Really, anything that you loved, whether or not it's related to economics or statistics, uh, that you think our audience would get a lot out of.
TH: Okay. So my bookshelves these days are filled with disaster books. Because, because I'm researching Cautionary Tales, I'm always picking up new books about things going wrong. And one that I've loved for a long, long time, is a book called Why Buildings Fall Down by Matthys Levy. And it's just beautifully illustrated and quite a nerdy discussion of all kinds of catastrophes where we have, we have buildings or bridges or other structures fell down and why they fell down. All the different ways in which they can fall down. Uh, and i- i- it's a great read, it's fascinating. It's easy to read. It's got these beautiful, uh, that little hand illustrations.
Uh, but I also felt that I learned a lot about the financial crisis of 2007, 2008 by reading, having read this book a couple of years before. Because it, it made me realize that the reasons that things go wrong is not just because somebody's, somebody had the wrong theory. So engineers understand why buildings stay up and why they fall down. And it's often because the world is just a complex place. And stuff happens that you didn't anticipate. There were details that you overlooked. Uh, the incentives are, are misaligned.
So lots of different reasons why things can go wrong. And starting with a, you know, setting where everything is basically pretty well understood. Like we know why concrete pillars support weight, start with that, and still and explore all the different ways in which things fall over. It's a great launching pad for understanding why systems fail in general. So Why Buildings Fall Down, it's a great book.
CG: Tim Harford, thanks so much for being on the New Bazaar. I tremendously enjoyed this conversation.
TH: Well, I, I love the podcast. I'm so honored to be on it. I really don't feel I should have made the cut but thank you. And, uh, I will continue to be a loyal listener.
CG And that's all for today's show. But very quickly before you go, executive producer Aimee Keane and I have a little holiday treat for you, the listener. Tomorrow, we are gonna run a part of our conversation with Tim Harford that you did not just hear. It's the part specifically where we chat with Tim about three of the stories that he has written about or done podcasts about. And then we discuss the economic lessons of each of those stories. So check your podcast feed again tomorrow for a bonus episode.
The New Bazaar is a production of Bazaar Audio for me and executive producer Aimee Keane. Adriene Lilly is our sound engineer. And our music is by Scott Lane and DJ Harrison of Subflora Studio. Please follow or subscribe to The New Bazaar on your app of choice. And if you like today's show, please leave us a review or tell a friend. If you wanna get in touch, I'm on Twitter, it's @cardiffgarcia. Or you can email us at hello@bazaaraudio.com Thanks so much for listening, and we'll see you tomorrow, actually.
Tim Harford is the author of numerous terrific economics books and the host of two great podcasts: “Cautionary Tales”, about what we should learn from big mistakes; and “More or Less”, about statistics. He also writes columns and essays for the Financial Times.
And what sets Tim apart in all these different mediums is his exceptional storytelling. When it comes to telling economic stories – stories that are meant to grab your attention, and keep you in suspense, and then ultimately land on a message or a lesson that really stays with you – I’m not sure there’s anyone better.
And Tim’s latest book, The Data Detective, is no different: there’s a lot of great stories in it. But what I really loved about it is that it’s very much also about the craft of storytelling itself. And that’s what today’s conversation with Tim is also about: What are the ingredients of a captivating story? Why is it that stories are so necessary for fighting back against misinformation? (Why aren’t the facts themselves enough?) And how do you wield the power of storytelling responsibly?
Links from the episode:
Cautionary Tales, hosted by Tim Harford, from Pushkin Industries
"The Problem with Facts", by Tim Harford
The Data Detective, by Tim Harford
