“Jane Austen, Game Theorist”

Stephen J. DUBNER: Hey Levitt, what is your, your personal favorite Jane Austen book?

Steve LEVITT: [LAUGHS]

*      *      *

Steve Levitt, my Freakonomics friend and co-author, teaches at the University of Chicago — in the economics department though, not the literature department. So when I asked him about Jane Austen…

LEVITT: Oh, man… I… didn’t she write a book… Did she write a book called Emma?

I wasn’t so surprised that the conversation kind of stalled out. Now, why would I ask an economist about Jane Austen, a novelist, who died in 1817?  Because today’s show is about game theory.

DUBNER: All right, Levitt, define game theory for me.

LEVITT: I would define game theory as the study of the strategic interactions between a small number of adversaries, usually two or three competitors.

So that sounds pretty simple, doesn’t it? Levitt has written several papers that involve game theory — mostly papers about sports and gambling and cheating, things like that. So how does it actually work? Well, here it gets a bit more complicated…

LEVITT: Yeah. So game theory, the promise of game theory… Or… Or…

LEVITT: So one of the predictions of game theory is that when you are in — oh well, let’s see. O.K., wait a minute. Wait, let me say again.

LEVITT: I would describe game theory as a mathematical formalization. O.K, O.K. let me start over. So…

DUBNER: Yeah, actually, you know what…

LEVITT: Should I talk? Actually ,you know what? Wait, let me… Let me talk about my frustration with game theory and then I’ll go back and say that I actually have written papers that game theory does apply.

DUBNER: Yep, yep.

LEVITT: So my applications of game theory, and there are a handful of them, have essentially all been to sports. Really my … sorry.

LEVITT: So, so my …

LEVITT: So my uh … uh, let me say it again …

LEVITT: Now, there are very particular predications that theory make about how pitchers show mitch … mitt …

LEVITT: There are very particular predictions about how pitchers should mitch.. mix their pitches.

LEVITT: Let me start over. Let me … Let me talk … Let me just like think differently about it.

DUBNER: O.K.

LEVITT: So when a pitcher sometimes throws fastballs, and sometimes throws curveballs, it must be the case that in the end, the pitcher must be indifferent between whether guy … God, you know game theory sucks so bad because its so hard.

DUBNER: [LAUGHS]

LEVITT: I mean, it’s really … I mean, because everything is backwards in game theory. I don’t even think it’s worth talking about, because like the predictions are just — they’re just impossible to describe without going into what equilibria is. So I don’t know how Michael does it.

Did you catch what Levitt said, right there at the very end of his, eloquent description of game theory?

LEVITT:  So, I don’t know how Michael does it.

Michael? Who’s this Michael?

Michael CHWE: My name is Michael Chwe, and I teach here at the political science department at U.C.L.A.

Michael Chwe — that’s C.H.W.E. — is an economist by training. Before U.C.L.A., he taught at the University of Chicago. That’s how Levitt knows him.

CHWE: And my research is on game theory, which is a mathematical subject, its applications to social movements, and macroeconomics, and violence, and this latest things is about its applications maybe to literature.

Game theory, as Michael Chwe sees it, is about thinking strategically, making conscious decisions, and making those decisions based on how you anticipate someone else responding to your decision. Think of a decision tree, with a lot of branches.

DUBNER: So if you would, tell us the name of your most recent book, then.

CHWE: Oh, yeah. Well, it’s Jane Austen, Game Theorist.

DUBNER: I have to say, that’s one of the best book titles I’ve heard in a long time.

CHWE: Oh, [LAUGHS] thank you so much.

DUBNER: It’s to the point, and it’s … it’s lovely and weird all at the same time.

CHWE: [LAUGHS] Thank you.

DUBNER: So Jane Austen Game Theorist, and tell me how the idea took root in your brain?

CHWE: Well, I mean, I was watching movies with my kids now they’re older, and we used to watch movies all the time together. And we watched Clueless, and I just thought it was a funny movie. And it was all about manipulation. And that’s with Alicia Silverstone.

DUBNER: Based, we should say, on a Jane Austen novel. Adapted.

CHWE: Exactly, it’s based on Jane Austen, and for a lot of people.

DUBNER: From Emma.

CHWE: Yeah, from Emma. And a lot of people feel it’s actually the best, in terms of the spirit, of the novel, the adaptation of Emma, which I kind of agree with.

Now, remember, Emma is the one Jane Austen book that Steve Levitt could name.

LEVITT: I have never read a Jane Austen book. The closest I’ve come to reading a Jane Austen book is I used to love the movie Clueless, with Alicia Silverstone. I watched that over and over. And I think that was actually one of her stories.

So academic economists love Clueless — who knew? And why might that be? Well, Clueless, adapted from Jane Austen’s Emma, is about a young woman who is constantly scheming to set people up romantically. This kind of scheming, it turns out, is not unusual in an Austen novel:

CHWE: Austen books are about many things, but really one of the major themes is the whole idea of meddling, manipulation, and scheming.

But Michael Chwe’s point isn’t just that Jane Austen is doing all this in her novels – Emma and Pride and Prejudice and Mansfield Park and Sense and Sensibility. His point is that she’s doing it intentionally – that Jane Austen was a game theorist. And this is noteworthy because she was essentially writing game theory more than 100 years before there was game theory.

CHWE: I guess, you could say if you’re looking historically that game theory was kind of part of the Cold War ethos in a way. I mean, a lot of early game theory was developed at the RAND Corporation, John Nash and Lloyd Shapley were very heavily involved in the RAND Corporation. And the RAND Corporation, of course, was a quasi-military think tank set up and funded by the U.S. government. You know, idea back then was that if we somehow modeled things like, you know, nuclear deterrents, or a ship being able to go after a submarine or something like that then we’d have some sort of advantage of the Soviets. But I’m not sure if it gives you a huge advantage. I think the thing it’s useful for is it simply gives you some clarity. It makes you think about things in a more clear way and brings up in some sense problems, which you might not have anticipated. So, that’s what I think game theory does for us generally. It doesn’t necessarily teach us how to save the world’s problems or anything, but rather, it sort of brings up interesting new wrinkles and interesting problems that you hadn’t thought about.

DUBNER: Yeah, I think a skeptic might say that game theory is a bunch of heavily educated, particularly mathematical people writing down complicated “elegant” formulas to “prove” what every small-business person or housekeeper or anybody has known their entire lives. Can you dispute that?

CHWE: To some extent, it’s true. I mean, people act strategically all the time.

DUBNER: Now, you write a few times throughout the book different versions of a similar statement. I want to run one or two past you and just challenge you to defend the thesis essentially. So you write that, “Austen consciously intended to theorize strategic thinking in her novels.” Another time you argue that Austen, “explicitly intended to explore the phenomenon, not of game theory per se, but of strategic thinking.” So persuade me first of all, Michael, that you do have explicit proof that this was an intention in her writing and not a case of confirmation bias where you, the game theory scholar, goes back to her writing and says, oh, this matches up very well with what I call this. But persuade me that she actually intended to do that.

CHWE: Well, I don’t have a smoking gun kind of letter from Austen saying, “Hey, I’m interested in strategic thinking. But a lot of people have asked me this. Obviously, Austen is interested in scheming and meddling. But saying that she’s explicitly interested in this as a theoretical subject is a different thing entirely. But that’s kind of a step up from that. So what I’ll say is that there are lots of little parables, or little asides in the novels, which don’t have anything really much to do with the plot or anything. You could just take them out and no one would care, but they do seem to be little explicit discussions of aspects of choice and aspects of strategic thinking. So, like, for example like in Pride And Prejudice, so like, you know, the very first manipulation is kind of what gets the whole novel started is, so you know the Bingleys’ come into town and so the Bennet family has five unmarried, you know, daughters, and that’s kind of a huge problem. So, Mrs. Bennet is super focused on getting her daughters married, and for obvious reasons.

It’s not like they can get jobs or anything. If that is the main way, you could become either a governess or you could get married. That’s basically it. So the very first manipulation is Mr. Bingley shows up with his sister and they rent out Netherfield, which is this estate nearby. And so Mr. Bingley’s sister invites Jane to come for dinner. And the first manipulation is Mrs. Bennet says, “Well, you’ve got to go on horseback.” And people say … The daughters say, “Why? Why horseback? Shouldn’t she take the carriage?”, and Mrs. Bennet says, “Well, it’s going to rain and if she goes on horseback, it is very likely that they will invite her to stay the night, and hence, she’ll get to spend more time.”

So, right from the beginning we have a manipulation. And that was — it seems kind of silly, but you have to play for keeps. This is a big deal. If you know, if somebody marriageable is nearby and you have a chance to spend 20 more minutes with that person, you’ve got to go for it. It’s … life or death is not too strong a way of thinking about this. So that manipulation gets started and it actually works too well. Jane gets wet in the rain and falls sick and spends several days with Mr. Bingley. And it kind of works. And, so in Pride And Prejudice Mrs. Bennet is not a very sympathetic character, and she seems to be very foolish, but if you look at what she accomplishes it is pretty good. She gets Jane married and she even sort of incentivizes Lydia — Lydia is the younger sister, who in a very sort of crisis-y kind of way, she runs of with Wigham without being married, which is a scandal. But I argue in the book that maybe she does that because she realized that the only way she can get some money in her marriage is to marry somebody who is not necessarily super committed to her so to create kind of crisis situation so the richer members of her family will then solve the problem for her. And that’s what happens.

Coming up ,on Freakonomics Radio: Back in 1813, Jane Austen was encouraging young women to think strategically. So who’s doing that now, 200 years later?

CHWE: I read in Sheryl Sandberg’s book that I should do this, and Sheryl Sandberg is asking me to ask you for a raise.

And later: Steve Levitt talks about why game theory is, for people like him, a big fat disappointment.

LEVITT: I think many economists as game theory got introduced had that same feeling about how game theory was going to open up the world to economists. And I think in the end that kind of wondrous promise was never fulfilled.

ANNOUNCER: From WNYC and APM, American Public Media: This is Freakonomics Radio. Here’s your host, Stephen Dubner.

Could the novelist Jane Austen be the godmother of game theory? Michael Chwe, an economist who teaches in the political science department at UCLA, says the answer is an emphatic yes — even though Austen died more than a century before game theory became game theory.

CHWE: It used to be thought that Austen kind of acted alone and wrote in her room and really didn’t talk to anybody else. But there’s a lot more evidence now of how she was definitely part of the intellectual world. She read Adam Smith, she read Hume, that kind of thing.

DUBNER: So it’s interesting. Here’s one line that Jane Austen wrote before, this is from a … Sorry, let me just make sure I know what this is actually from. Yeah, O.K. So this is from — well, it’s called A Memoir … I don’t know this book, A Memoir of Jane Austen, by …

CHWE: O.K. Oh, it’s by one of her nephews. 

DUBNER: Austen-Leigh. Is that a nephew?

CHWE: Yeah, exactly right. He was like the first person who really kind of memorialized and gave a biography of Austen. So he kind of — he’s like — the beginning of Austen-mania started with him.

DUBNER: I see, oh very good. O.K, so he writes that she wrote before she began Emma that she wrote, I guess, she wrote in a journal or a letter, “I am going to take a heroine,” referring to Emma, “whom no one but myself will much like.”

CHWE: [LAUGHS]

DUBNER: So she sounds like an economist there to my mind, because economists are quite proud of the fact that they can make dispassionate argument about anything, you know, labor markets, love markets and so on. And so, I guess what I want to know is how much you know or have thought about, or have been able to learn about the kind of person she was and what her family situation was like and how that influenced the way that she thought. You mention, you know, she read Smith and Hume and so on. We know she was from gentry, right kind of lower gentry level, is that right?

CHWE: Right. I mean, there are people who spend their entire lives doing this and reading her letters and trying to get more into it. And I haven’t done that as much. I mean, I would simply say that, these idea were floating around, but also at the same time like in the book I argue that a lot of the slave folk tales talk about game theory. And it’s not like slaves were in touch with David Hume or anything like that, right? So I’m arguing that actually a lot of these things are kind of close to common sense. It’s just kind of like elaborated common sense. It just like a lot of people say hey game theory sounds a lot like common sense. And I say, yeah, it actually does. It is common sense.

DUBNER: Common sense perfected. Yeah.

CHWE: Yeah, exactly. That’s a good thing.

DUBNER: All right, so give me another example or more evidence that Jane Austen was thinking like an economist.

CHWE: Well, I mean, the one example I give sometimes is in economics we have this idea of opportunity cost. When you take an action you evaluate the action not just in terms of its payoffs, but also how good it is relative to some other thing. Right? So, for example, if I go to graduate school, the opportunity cost is the money I would have made had I gotten a job. So, this is an interesting little economics concept, it comes from thinking about choice. In other words, you always choose between not just A and nothing, but A and some other alternative, which is a real alternative, B.

O.K, so one little thing in the novels, is in Emma, Emma and her friend Mrs. Weston are talking about Jane Fairfax. Jane Fairfax is another young fashionable woman who’s in the area, in the neighborhood. And they say, “Well you know, she spends a lot of time with Mrs. Elton who’s kind of a silly person, not very thoughtful.” So why is she spending so much time with Mrs. Elton? And Mrs. Weston says, “Well, before we condemn what she chooses, we have to consider what she quits.” In other words, we have to consider the fact that at home, if she stayed at home she just has to hang out with aunt and her granddad. So in other words, to think about her choice we have to think about what she would have done otherwise. And again, there’s no reason for that to be in the plot. There’s no reason for that little discussion to be there.

DUBNER: Now one thing that’s particularly interesting about Austen is she was obviously a female novelist at the time when there weren’t so many novelists at all and being a female novelist was not the most natural or common thing in the world. And additionally, most of the characters that we really care about are females themselves. So talk for a minute about game theory and girls and young women the way that they use it, or Austen uses it for them in her novels.

CHWE: Well, I mean, the book kind of makes an argument that women are generally better at strategic thinking in the real world. So there’s some evidence from psychology, like a classic thing is theory of mind tests, like are you able to put yourself in the mind of another. And can you tell somebody’s emotions just by looking at their eyes. And women in general tend to be better at this than men.

DUBNER: So let me ask you this question, and this is probably unanswerable, but I want to know what you want to say about it. Let’s assume that women are better at what we call strategic thinking than men. Do we have any idea of whether the why in that is something to do with genetics or something to do perhaps with the fact that women are more often in a position where they need to learn to be strategic thinkers because they have historically had a lot less power in a situation like that?

CHWE: Yeah, there might be a genetic explanation. I’m not going to rule that out, but I would, the book stresses the second one, which is exactly that, people who are in situations where they don’t necessarily have a lot of power, that’s exactly the kind of person who needs to think strategically. And so in the book I call this, sort of, game theory is one of the original weapons of the week. I mean, it’s one of the original ways of getting ahead if you don’t have a lot to start out with.

And in some sense if you already have a lot of power you don’t have to think strategically because everyone else is already doing what they’re supposed to do, you’re happy with that. I have a friend, Marek Kaminski, who’s at U.C. Irvine, and he was in a Polish prison. He’s Polish and he was put in prison because of democratic activities he did. And he said that in prison, prison totally trains you to act hyper rationally because in prison it’s possible the one right move in the right spot could give immense consequences, like get you out early, or transfer you to a different cell. So he says the world of prisoners is hyperrational. So it’s exactly this kind of thing, because if you don’t have a lot of power you need to, you know, use whatever agency you have in the best possible way, and strategic thinking is absolutely required.

DUBNER: You write that young women of the day often learned not necessarily the social cues and mores, but the means of strategic thinking through novels, because, you know, they were one of the few shared forms of public communication, I guess, at that time. What about now? You have children. Do you have daughters?

CHWE: Yeah, I have a son and a daughter. I have a son in college and my daughter’s in high school.

DUBNER: O.K. So I’m curious to know from your perspective where your daughter has learned strategic thinking. I gather it’s not from a Jane Austen novel.

CHWE: That’s interesting. I mean, I think, how do you learn it? I guess you learn it from, well, just ordinary interactions. So, like in the book I say Catherine Morland, she’s like 16 or 17, and she learns, you know, just for example, she learns how to make your own decisions by basically going along with everyone else for a while and then realizing, “Hey, going along with everyone else isn’t really working out for me, so I’ve got to make my own decisions.” So some of it is through real life. But there is this kind of shared gossipy culture. You know, like the other day I read this article about, you know, Sheryl Sandberg’s Lean In book. So, Sandberg says in her book, if you ask
for a raise, don’t go in to your boss and say, “Hey, I need a raise.” Say, “Actually, one of my superiors, i.e. your bosses’ boss told me I should come and ask you for a raise.” So it’s a very specific strategic thing.

What’s great about it is it takes the onus off of you, right? And it says it’s not about me asking for a raise, it’s someone else telling me I should get a raise. So, you know, kind of just don’t blame me. And so what’s interesting is this article said not only are women asking for raises in this way, which is a nice strategic thing, but they actually say, “Now, I read in Sheryl Sandberg’s book that I should do this, and Sheryl Sandberg is asking me to ask you for a raise.” So what’s going on there is, well again, I think Sandberg’s book is kind of like another shared strategic culture among women. It’s part of a long tradition, which Austen’s a part of, which occurs among people’s conversations everyday, which is a way about how to go through life, and how specific techniques really work. And why not try them, and we share these things among each other. And you know, it’s kind of like a shared strategic culture.

Let’s get back to Steve Levitt, my Freakonomics friend and co-author. As you heard earlier, Levitt is not really the biggest fan of game theory.

LEVITT: God, game theory sucks so bad because it’s so hard.

Part of this disillusion may be because game theory was supposed to be the next big thing in economics. And then it wasn’t.

LEVITT: When I first got introduced to game theory, it seemed to me like it might be the answer to everything important. I think many economists as game theory got introduced had that same feeling about how game theory was going to open up the world to economists. And I think in the end that kind of wondrous promise was never fulfilled. And the difficulty is that game theory really only applies to a narrow set of problems. That’s a set of problems where there are just two or three, or a very small number of actors. And it really does much better when either the game that is being played is repeated an infinite number of times, the same game is played exactly over, and over, and over to infinity, or it’s played precisely once. It turns out that in the middle ground of there being a handful of participants, or a handful of plays, game theory doesn’t often do such a great job of solving our problems. And in the real world, there just aren’t very many problems that end up being the right kinds of problems for game theory.

DUBNER: Let me just ask you this, Levitt, what does it take to be good at game theory?

LEVITT: There are two things that are important to doing well in strategic settings. And the first one is knowing enough and being skilled enough to put yourself in the shoes of the other person. So you cannot do game theory unless you can say, “If I do this, she will do that. If I do that, she will do this.” Because that is so fundamental to game theory that if you aren’t in the habit or don’t have the ability to understand how someone will react, you have no hope whatsoever. The second trait, which is valuable, is to be able to look many steps into the future. So you can be only so good at game theory if you can think to yourself, “If I do this, then he does that.” Really good game theorists, the most skilled ones will say if, “I do this, then he’ll do that. Then I’ll do this, then he’ll do that. Then I’ll do this and he’ll do that.” And that’s kind of the difference between a really good chess player and a not so good chess player is being able to see down the road much further.

DUBNER: But you and I have talked a lot over the years about how the future is essentially unknowable at least in certain realms and to certain degrees. So how do you get good at, or better at knowing what someone else, whether it’s a collaborator or an opponent will do in the future.

LEVITT: I think the answer to that question must be the answer to every question that involves getting good at something, which is feedback, trying and experimenting and learning whether or not your insights were correct or incorrect, and then updating your behavior as a function of that. I think the only way to learn is to practice, and to practice in settings in which you get good feedback about whether you’re right or you’re wrong. But there are substitutes for practice, and that’s reading about stuff. So you don’t actually have to do stuff if you can read about smart people and how they’ve done things that is another way to learn.

Huh, that’s an interesting idea. Learn to do by reading about stuff. By reading about how smart people do things. By reading an author like — hell, I don’t know — Jane Austen.