As I’m sure all astute Duck readers are aware, today marks a critical day in the US House and Senate – if no deal is struck today on a spending bill, the US government will shut down at one minute after midnight on Tuesday morning. The issue at the heart of the controversy: a series of amendments to the spending bill that concern the Affordable Care Act (so-called “Obamacare”). In general, House Republicans are in favor of the amendments; Senate Democrats are against the amendments. So, both sides are holding firm to their stance on the amendments in hopes that the other side caves in before tomorrow. What are the likely outcomes of this situation?
Editor’s Note: This is a guest post by Eric Grynaviski, who is an Assistant Professor of Political Science at George Washington University.
The recurring debate on this blog has centered on some of the bigger themes about the relationship between rational choice theory and game theory. I argued in an earlier post that when one focuses in on the specific logic of rational choice theory and/or game theory, getting away from its abstract characterizations, there are some similarities about the way it understands society with alternative approaches.
In this post, I want to focus on a more specific issue, which is how to understand the relationship between pragmatism and rational choice theory. Pragmatism has recently been used in IR in two ways. On the one hand, pragmatism has been invoked to justify a particular image of science, usually (but not always) one that is post-paradigmatic or pluralistic. Others have concentrated on the pragmatist contributions to IR theory or ethics.
This post concentrates on pragmatism as social theory. The early pragmatists—Dewey and Mead in particular—were very interested in questions such as logics of action that are at the heart of modern-day IR theory. I want to argue that there are a lot of similarities between pragmatism and rational choice theory, providing at least one via media between sociological and economic approaches that has been unexplored to date.
There are significant differences though between the ways rational choice theory and pragmatism tend to model learning and reasoning. The nub of the problem that this post concentrates on is uncertainty. Rational choice theorists tend to describe some set of possible worlds (a state space) over which agents assign probabilities. Reasoning and learning usually involves agents changing those probabilities in response to new information. Pragmatists tend to be interested in why possibilities become possible or become impossible; they are interested in how states enter and leave the state space and not how probabilities are assigned.
This post concentrates on two issues. Conceptually, is a rational model of action compatible with a pragmatist theory of action? And second, what are the differences in their treatments of certainty. Continue reading
Editor’s Note: This is a guest post by Eric Grynaviski, who is an Assistant Professor of Political Science at George Washington University.
There has been a debate on the Duck lately about the meaning of rational choice theory and game theory, and how it’s different from varied alternative approaches (here, here, and here). I wanted to offer a different interpretation than Arena and Jackson. Both give pretty orthodox interpretations, where game theory treats human agents as economic agents interested in maximizing their utility. I wanted to offer a sociologically richer interpretation, concentrating on the idea of how agents react to strategic interdependence.
At the heart of the logic of n-person games is the idea that the best move I make can be affected by what others will do. This is the idea of strategic interdependence. Nexon writes about strategic interdependence that it’s no big deal: “Strategically interdependent preferences of individual actors are not the same thing as a collection of relationally embedded actors.”
The way game theory deals with strategic interdependence gets us into relationally embedded actors. It is a really big deal sociologically speaking. Continue reading
So it increasingly looks like the inter-Korean Kaesong industrial zone is closed for good. (The Wikipedia write-up is a pretty good quick history of it.)
The zone was set-up during the Sunshine Policy period (1998-2007). It was to do 3 things: 1) Lead to some liberal-capitalist spill-over in the North, 2) Expose regular North Koreans (the workers in the area) to regular South Koreans (the managers and staff), and 3) Generally provide some inter-Korean cooperation that might hopefully reduce larger tensions. A resort area in North Korea (Mt. Kumgang) was also opened along these lines in the Sunshine period. Broadly the idea was along the lines of liberal explanations for the Soviet Union’s changes in the 1980s: the Helsinki Accords and CSCE opened the USSR to the outside world, and the inflowing liberalism slowly changed attitudes that eventually helped wind-down the Cold War. Unfortunately, none of this seems to working in the NK case.
Below, Scott Weiner argues that Carly Rae Jepsen’s song “Call Me Maybe” is an illustration of the dynamics of standard game-theory models, specifically the prisoner’s dilemma and stag hunt. Weiner assumes that Jepsen is a rational actor, that both Jepsen and her beau are better off being together than being apart or with different partners, and that Jepsen is rationally choosing to communicate her availability to facilitate their coming together. I share these assumptions, but as I demonstrate Weiner misses the key points of the song. If, as Weiner suggests, both Carly and the boy are better off together than apart, then why signal that “this is crazy”? And why is the song called “Call Me Maybe” instead of “Call Me Right Now So We Can Be Together”?
The answer is that Carly is trying to communicate that, despite her forward approach to the boy, she is nevertheless suitable for him. Sometimes, disclosing more information hurts rational actors, and for Carly to disclose that she is interested in the boy after having just met him could signal to the boy that she is an undesirable partner—not just because of old-fashioned notions (“she’s not wife material“) but also because an aggressive partner of either sex might not be interested in a long-term relationship (Hall and Oates, 1982).
So we are left with a puzzle. If Jepsen is rational and can assume her potential partner is as well, why pursue a strategy that both stresses her availability (“call me!”) while highlighting her ambiguity (“maybe?”) and stressing that the situation is causing her to behave in an unusual way (“and this is crazy”)? The answer lies in the fact that dating is a game played under asymmetric information, which changes the dynamics of the interaction in ways Weiner does not appreciate. I provide an informal treatment below.
Assume there’s some distribution of types of potential dating partners in the world, “worthy” and “tragic.” (We assume that the dating game is multiple-shot; as is well understood, one-shot romantic games have dramatically different properties.) The preference of each player, worthy or tragic, is to find a worthy partner and to avoid ending up with a tragic partner. Worthy partners would rather be alone than with a tragic partner; tragic partners would rather be with a tragic partner than alone.However, although every player knows his or her type (that is, whether they are themselves tragic or worthy), they can’t know with certainty whether other players are. Consequently, players who advertise themselves as worthy may be lying, and there’s no way to tell in advance.
How, then, for worthy partners to advertise themselves as being worthy? As Schelling and others would point out, there has to be some sort of credible signal. This, however, is likely to be reticence, since tragic partners are made much better off by being with anyone than by being with the right partner. Consequently, the dating scene is likely to be made up of tragic partners pretending to be worthy ending up with each other. (Game theory is often realistic that way.) This is a perverse equilibrium: The only players left on the scene are the ones who shouldn’t be dating anyone, because all the worthy partners know that trying too hard puts off other worthy partners.
Let’s assume, however, that Carly and her boy are both worthy. If Carly comes off too strong, then the boy may assume that she is tragic. So she instead engages in signaling by saying that she’s not normally this way, that the situation is highly unusual, and that she’s putting off all the other boys who are interested in her to talk to the boy–all signals that she is interested but not tragic.
Unfortunately for Carly, the ploy is unlikely to work if the boy is a worthy partner. While Weiner does not provide an independent assessment of how likely Carly and her object of attraction are to end up together, his analysis suggest that they will be happy together because they are better off together. Alas, my analysis suggests instead that all such posturing will be dismissed as merely cheap talk.
This is a guest post by Scott Weiner, a PhD student in Political Science at George Washington University.
One of this summer’s most popular hit singles is “Call Me Maybe” by pop artist Carly Rae Jepsen. In the song, Carly attempts to score a date with an attractive male by giving him her number and asking him to call her in order to set up the outing. This strategy is eventually successful, and while the male “took his time with the call,” Carly “takes no time with the fall.” This outcome is puzzling given that existing accounts of the scenario might predict a sub-optimal outcome given Carly’s strategy. Why does Carly Rae Jepsen give the boy her number despite her own realization that “this is crazy?” Why does Carly Rae Jepsen tell the boy, ambiguously, “call me, maybe” when her preferences are not at all ambiguous given that she very much wants him to call her? How can scholars understand the successful outcome of this strategy?
Existing literature understands the basic scenario presented in “Call Me, Maybe” as a prisoner’s dilemma. In the prisoner’s dilemma, two rational actors who cannot communicate with each other are given a choice of cooperation with each other or defection, with a system of rewards and penalties for each:
In the basic prisoner’s dilemma, the optimal strategy is to defect since the cooperation of the other actor cannot be guaranteed. Each actor’s payoff will be better by defecting regardless of the choice of the other player. Since Player A cannot guarantee the cooperation of Player B she will choose the best course of action for herself regardless of B’s choice.
For the purposes of this model, we can assume Carly Rae Jepsen is a rational actor. She begins the song with the words “I threw a wish in a well / don’t ask me I’ll never tell.” This indicates a clear set of preferences. The fact that she will not reveal her wish under any circumstances indicates that these preferences are constant throughout the game. Carly also sets up a ranked order of preferences, noting “I’d trade my soul for a wish / pennies and dimes for a kiss.” This monetization of kisses indicates her ranking is in fact quite sophisticated.
However, assuming the boy is a rational actor as well (which Carly does) the prisoner’s dilemma would predict that her optimal strategy is to defect. Since she cannot guarantee the boy will call her, the prisoner’s dilemma predicts she should not give him her number, and that her actions are, in fact “crazy.” What accounts for not only Carly’s actions, but also the success of her strategy? To answer this question, we must look beyond the constraints of the prisoners dilemma. Other models may in fact lend more explanatory leverage on the issue.
I. A Shadow of the Future
One of the most important rules at play in a classic prisoner’s dilemma is that it is a one-shot game. However, if the game is played over and over with the same actor, this is known as an “iterated prisoner’s dilemma.” In this case, since the game is repeated, each actor will have to live with the consequences of his actions after the first round is over. This added condition is called the “shadow of the future.” When a shadow of the future is present perpetually (i.e. the game does not have a set end point), the optimal strategy ceases to be one of defection and instead becomes a “tit-for-tat” strategy, in which the actors try to mirror each other’s actions (see Axelrod’s “The Evolution of Strategies in the Iterated Prisoner’s Dilemma“)
Carly opens the game by giving the boy her number, which is cooperation. Since if they were to date the game would repeat without a definite end-point, Carly calculates that it is in the boy’s rational interest to call her. Until the point that either Carly or the boy defect from the game, cooperation is the optimal strategy according to the model.
However, the reality is not quite so simple. Rationally, Carly should signal every intent to cooperate to the boy in order to maintain her credibility. Yet she deliberately tells him “Hey, I just met you / and this is crazy.” What explains this puzzling signal?
II. Signaling Intentions In The Stag Hunt
Carly’s predicament could also be explained via a model known as the stag hunt. Originally developed by Jean-Jacques Rousseau, the stag hunt involves two players who get a small payoff from hunting two rabbits separately but a large payoff from hunting one large stag together. Hunting stag requires a different weapon than hunting rabbit, however, and the weapon choice of the other player is unknown.
In a sense, Carly and the boy in question are in a sort of stag hunt. We assume for the purposes of the game that both Carly and the boy would prefer to go on a date over not going on a date (“Before you came into my life / I missed you so bad”). However, they also do not want their time wasted by trying to score a date with someone who is uninterested in going out with them. We can model the payoff structure of the game as follows:
As the matrix reveals, there are two equilibria in the game, but one has a higher payoff than the other. When such a payoff structure exists, actors will try to communicate their intention to cooperate (ie, go on a date) in order to try to induce cooperation from the other party. Communication is a highly theorized area of international relations, which involves signaling capability, resolve, and credibility. How can we understand Carly’s communication in this regard?
Carly’s statement “Hey, I just met you / and this is crazy” is an attempt to communicate both intentions and resolve. In particular, both statements are intended to highlight the costly signals Carly is giving of her intentions. Were Carly not interested in the boy, giving him her number after having just met him, an admittedly “crazy” action, would incur significant costs. By doing so regardless, Carly is communicating that she is in fact interested in having him call her. Her willingness to challenge social norms is an attempt to communicate resolve, especially given communication difficulties implicit in the situation at hand (“It’s hard to look right / at you baby”). That is, she is in fact interested in the boy and does in fact want the boy to call. Carly supports this signaling regime by noting that “all the other boys / try and chase me” a statement that she is committed to exclusive cooperation with the boy at hand.
The addition of the word “maybe” at the end of her signal is a tactic designed to highlight the choice which the boy now has to make between calling and not calling. Schelling would categorize “maybe” as as a “trip-wire,” in which one actor sets up an automated series of events which the other actor will trigger with a certain action. Since the first actor, Carly, has already chosen a risky course of action and the decision is out of her hands, it falls to the boy to pursue a strategy with the lowest risk for himself. This also turns out to be the one with the biggest payoff for Carly as well. As it happens, the boy does eventually call, and both Carly and the boy achieve their Pareto-efficient equilibrium.
In conclusion, Carly’s strategy is actually a rational one given the payoff structure she faces in the given situation. While such an explanation cannot explain her decision in the music video to wash a car in 5-inch heels, it can explain her actions as the outcome of a rational strategy. Further research should examine the generalizability of the argument by accounting for critical cases such as “Payphone (explicit)” by Maroon 5 (ft. Wiz Khalifa) and “Wide Awake” by Katy Perry. Ultimately, such inquiry serves to provide scholars with a deeper understanding of the complex world of interpersonal relations as relayed through pop songs.
At the end of the show the contestants have to make one last decision over the final jackpot. They are each presented with two golden balls. One has “split” printed inside it and the other has “steal” printed inside it:
If both contestants choose the split ball, the jackpot is split equally between them.
If one contestant chooses the split ball and the other chooses the steal ball, the stealer gets all the money and the splitter leaves empty-handed.
If both contestants choose the steal ball, they both leave empty-handed.
It is similar to the prisoner’s dilemma in game theory, however, in this game the players are allowed to communicate.
Indeed, the communication is the interesting element of this particular play of the game:
Here is the “Golden Balls” situation using simple 2×2 game matrices:
In this game, steal is likely the dominant strategy. If you are certain your opponent is going to split, then it is superior to steal in a single play. You win. If you are certain your opponent is going to steal, then you are indifferent between stealing and splitting, though many people would likely steal just to avoid being made to be the sucker (thinking of relative gains).
Indeed, if we ignore cash values and make the sucker result the 4th-ranked payoff given the logic I’ve just provided about relative gains, then this game would then be a single-shot prisoner’s dilemma game. The dominant strategy is steal (defect). Obviously, preferences over outcomes should determine the strategy one employs in a game. Generally, however, simple game theory assumes utility maximization and the outcomes here are technically the same.
In any case, in this video from “Golden Balls,” player 1 (the man on the right in the brown shirt) has attempted to turn this situation into a different game — chicken, I think — by trying to add a perceived payoff that is worse than playing the sucker in a prisoner’s dilemma.
In chicken, the common story is two teenage drivers head directly for one another at high speed. If they both swerve (yield), this is the mutual split result. If only one swerves, s/he is the chicken and the other player wins. If both continue driving towards one another, they have a horrible accident.
Here, if player 2 selects steal with the knowledge that player 1 is definitely going to steal, then the total prize possible will be ZERO. However, if player 2 lets himself be exploited, then player 1 has dangled the (unenforceable?) promise of sharing the winnings after the show. Effectively, player 1 has attempted to transform the situation by creating the image of a shared victory even when the other player yields. It would be kind of like a fixed boxing match. The payoff comes after the participant takes the dive.
Generally, if one earns a reputation for selecting steal (never swerving) in the game of chicken, then no others will want to play this game with you because their best option is to split (swerve/yield). Why select the outcome that will assuredly result in a disastrous outcome? Unfortunately, one cannot earn a reputation for unyielding play in the first confrontation with an unknown player.
However, in his Introduction to Herman Kahn’s On Escalation, Thomas Schelling recommended that a chicken player should throw the steering wheel out the car window to signal a firm commitment to the steal (not swerve) strategy. Such a player has signaled to the opponent that the result is out of his hands. The best that can be hoped is to avoid disaster.
In this case, to influence Player 2’s choice, Player 1 has essentially communicated that he is tossing the steering wheel out the window.
Political scientists like fantasy role-playing games. But this does not mean they are simple nerds. They like a particularly elegant and sophisticated escapism called game theory. While it might seem obscure and overly complicated at first, you can grasp game theory and impress your political scientist friends with a very simple insight — game theory is just like Dungeons & Dragons.
Like Dungeons & Dragons, game theory players embark upon imaginary adventures in which they interact with others in situations never before seen in the real world. The game theorist operates as the dungeon master, setting up a stylized environment in which players cooperate and compete over some prize such as being elected, winning a missile crisis, or maintaining a fixed exchange rate system. He, always he, sets up the game tree or matrix that describes the actions that are possible at different moments in the adventure or campaign and gives them their powers like the ability to veto or escalate. The outcomes are based on probabilities, although game theorists do not use dodecahedron dice. And like Dungeons & Dragons, the outcome is of no consequence for understanding the actual world around us.
Game theorists believe that active use of imagination clarifies complex situations and concepts. They point out that simple games like Chicken, the Prisoner’s Dilemma or the Battle of the Sexes are apt metaphors that help us understand any number of strategic interactions in the real world. It is unclear how game theorists came about the inspiration for those games, as no game theorist has ever talked to a girl, much less been forced to choose between a romantic beach or mountain holiday. Their interactions with the opposite sex generally revolve around Japanese anime. And they aren’t exactly James Dean or hardened felons either.
Bargaining must have broken down due to incentives to dissemble. Dragons do indeed have a hard time making credible commitments as they make the offense dominant.
Game theory is sometimes called formal modeling, which is an unfortunate term, as no game theorist has ever worn a tuxedo, ever. They are generally pleased to find some sweatpants at the bottom of the laundry basket without a stain. And game theorists avoid having their picture taken when possible because… well, there is the sweatpants, for one.
It is not known whether a successful career in game theory is correlated with prior experience as a dungeon master although there are clear signs that this might be the case. Game theorists and D&D players both spend considerable time in windowless rooms. And both try to avoid any contact with genuine empirical data, whether it be books or the reality that there are no sexy witches in the real world.
However, there are clear differences as well. Game theorists make considerable sums of money while most D&D players, even adult ones, still live in their parents’ basement. And game theorists have clearly lost all sense of mystery. Their players are all colorless automotons who cannot talk to each without fearing that the other is lying, much less fly. They are, however, remarkably capable of making precise estimates of probability which is kind of like magic.
It is possible that game theorists are fantasy enthusiasts who no longer possess an inner child’s sense of wonder. This is perhaps due to the crushing experience of interacting with other game theorists on a regular basis.Contrary to longstanding rumors, however, game theorists are not Satanists. It is all just good clean fun. Indeed their preferred environment, a godless dystopia of egoistic utility maximizers, suggests that they are much more likely to be atheists. It is also not possible to harm a game theorist with silver, although they do suffer great pain when their on-line access to the American Political Science Review is cut off.
Daniel Drezner asks whether there are any options for dealing with evidence of North Korea’s involvement on the sinking of the Cheonan besides diddling around in the UN Security Council with a resolution that China may well veto. He rightly suggests that this is essentially a game of chicken that the North always wins because it seems crazier and less predictable than most civilized states (true). He points out the conventional wisdom that war must be avoided at all costs because Pyongyang is poised to deal a devastating blow to Seoul (and Pyongyang, for its part, knows it would probably be defeated). He proposes that the international community not allow North Korea to participate in the World Cup.
That latter is not a bad suggestion. As Alegi and Bolsmann have documented, sports sanctions made a difference in ending apartheid, and a rash of new studies including this one make similar arguments. And as a human security analyst, I’m glad to see that the protection of civilians in Seoul is a top priority for those ruminating over how this crisis might develop.
However, as a human security analyst, I’m equally concerned with two things:
a) the protection of civilians in North Korea, where 30% of the population is starving, where 400,000 people languish in Soviet-style gulags, and there is a near-complete absence of civil and political rights) – something that can probably only be accomplished by significant changes in the political culture of the DPRK and
b) the long-term stability of the region, where the status quo seems to be based on “containing” a (crazy and unpredictable) North Korea – a move that may ultimately fail, with catastrophic later consequences for civilians in Seoul and elsewhere.
So I’m wondering if the real answer to Dan’s question about chicken requires rethinking the structure of the game. I don’t have enough expertise on the region to translate this into concrete recommendations but the way I would re-frame the question is like this:
What are the range of options (if any) for sending costly signals to DPRK that imply that South Korea might be readier to absorb the consequences of a land war than DPRK would be to absorb the likelihood of losing one? For example, since one of the key concerns is the vulnerability of Seoul to artillery fire, what measures if any could be taken by South Korea and the international humanitarian community to reduce the likely civilian casualties of a strike from the north on Seoul, thereby making the threat of massive casualties less crippling in such an event? Or, what measures could be taken by South Korea’s allies to preempt such a strike rather than waiting for it (which would probably be within the limits of the UN Charter regime as well as responsibility to protect doctrine given recent DPRK actions)? And are there any options that put improving the lot of North Korea’s own civilian population on the same footing as concerns about Seoul’s civilians or regional stability?
I ask these questions of Dan in my latest bloggingheads diavlog. In asking these questions, I’m not suggesting (or at least not meaning to suggest with any certainty) that all-out war on the peninsula is desirable (though limited strikes may be – I’d have to understand the force structure in the region better than I do). But my key argument is that behaving in any situation as if we think war is unthinkable gives the opponent all the leverage. If this is actually a game of chicken as Dan argues, how might the policy dilemma be framed in such a way that North Korea, who actually wants to avoid war, might start to believe that it’s not the craziest party in the equation anymore or the one with the least to lose?
Tim Hartford (whose blog at FT.com you really must read) discusses the results of a recent survey that suggest the answer is yes:
A recent survey by Yoram Bauman and Elaina Rose, two economists from the University of Washington, explains that in experiments, economics students are less generous, more likely to choose an unco-operative approach and more likely to accept bribes.
Bauman and Rose’s survey built upon an earlier study 30 years ago which demonstrated that “postgraduate students of economics were more likely than others to “free ride” in a laboratory game, effectively exploiting other players for their own benefit.”
I tend to agree with Hartford’s supposition that what is really going on here is that economists–as well as political scientists and sociologists–are simply choosing optimal strategies based on the game theoretic models upon which the laboratory experiments are based. Cooperation is not inherently a good strategy, but rather one that is determined by the structure of a game or experiment (e.g. what is the payoff structure of particular combinations of choices, is the game a one-shot deal or is it iterated, etc). Social scientists are trained in, and therefore comfortable with, game theory and the various structures and payoffs that exist. It is reasonable then to assume that if placed in an experiment that mimics those structures and payoffs they are more likely to play the most dominant strategies.
For example, if I recognize that the experimental situation is a one-shot, Prisoner’s Dilemma then I am going to defect rather than choose to cooperate. Why? Because the outcome depends on my choice as well as my fellow subject, and the structure of the game dictates that defection is the dominant strategy for both parties–why assume the other subject would choose differently, particularly given the risk I run of a huge loss if they don’t choose to cooperate? Now, if they game is iterated and neither of us know when we will stop having to choose to cooperate, the shadow of the future makes cooperation a more dominant strategy. As Hartford noted:
[P]erhaps the budding economists are not truly mean and selfish, but are simply showing that they have mastered their studies by producing the behaviour described in simple textbook models. Arguably, the students of economics are not doing anything sinister, any more than if they calculated the roots of a quadratic equation.
There is also the possibility that those that choose to enter postgraduate training in the social sciences are simply more jaded, cynical, or “realist” in their worldview. And while they may hold personal views that cooperation and selfless behavior are desirable and moral endpoints, their research and training illustrates to them that in many cases it can be unproductive (or, in some cases, counterproductive) to cooperate oneself without taking into account what others will do.
In World Politics yesterday we covered the Peloponnesian War, the Melian Dialogue, and the security dilemma as an introduction to realist theory. Students played a version of the 2-person non-iterated prisoner’s dilemma game developed by my former professor Robert Darst, with the winners receiving candy and the person with the lowest possible grade receiving an extra credit point toward their final grade. The students learned that the incentive structure in the game is a powerful causal variable affecting outcomes: when the game is structured so as to reward rational, self-interested behavior, cooperation becomes foolhardy, even if your intentions are noble. Realists would say this reflects the nature of the international system under anarchy.
Then again, game theory also predicts that if you change the parameters of the game you change the possible outcomes. The clip above from The Princess Bride demonstrates the basic idea of game theory, and also how changing the nature of the game is the best way to get what you want. But there’s many a slip between cup and lip – between manipulating perceptions within the context of the same parameters and changing the game itself. Unfortunately, realists are not optimistic about the latter happening unless a world government is established.