APPLYING GAME THEORY TO PRESIDENTIAL MISTAKES Except where reference is made to the work of others, the work described in this dissertation is my own or was done in collaboration with my advisory committee. This dissertation does not include proprietary or classified information. ________________________ Anida Wishnietsky Certificate of Approval: ________________________ ________________________ Anne Permaloff Carl Grafton, Chair Professor Emeritus Professor Emeritus Political Science and Public Political Science and Public Administration Administration ________________________ ________________________ Christa Slaton Bradley Moody Professor Associate Professor Political Science and Public Political Science and Public Administration Administration ________________________ George T. Flowers Interim Dean Graduate School APPLYING GAME THEORY TO PRESIDENTIAL MISTAKES Anida G. Wishnietsky A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn, Alabama December 17, 2007 APPLYING GAME THEORY TO PRESIDENTIAL MISTAKES Anida G. Wishnietsky Permission is granted to Auburn University to make copies of this dissertation at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights. ______________________________ Signature of Author ______________________________ Date of Graduation iii DISSERTATION ABSTRACT APPLYING GAME THEORY TO PRESIDENTIAL MISTAKES Anida Wishnietsky Doctor of Philosophy, December 17, 2007 (M.S., University of Southern Mississippi, 1984) (M.A., Webster University, 1976) (B.A, University of Miami, 1973) 188 Typed Pages Directed by Carl T. Grafton Game theory, based on the rational actor model (RAM), is a decision making theory that is useful for predicting economic decisions but is not reliable for predicting political decisions. Game theory might work better if a utility function could be accurately determined and applied to the players of the game. Three other decision making models (organizational behavior model (OBM), governmental politics model (GPM), elitist actor model (EAM)) are used to develop the utility function to better utilize game theory to explain and predict decisions. Four major presidential mistakes (Kennedy?s Bay of Pigs, Johnson?s escalation of Vietnam, Nixon?s Watergate, and Clinton?s Lewinsky scandal) were used to test the theory. The results demonstrate the usefulness of this method to explain, but not predict political decisions. iv ACKNOWLEDGMENTS The author would like to thank Kenneth R. Barber for his love, support, and gentle nagging, which kept this study on track and progressing toward a timely conclusion. v Style manual or journal used: Chicago Manual of Style Computer software used: Microsoft Word vi TABLE OF CONTENTS LIST OF TABLES........................................................................................................... viii CHAPTER 1. INTRODUCTION ........................................................................................1 CHAPTER 2. REVIEW OF THE LITERATURE ..............................................................5 The Present Study ..................................................................................................20 CHAPTER 3. METHODOLOGY .....................................................................................21 Research Question .................................................................................................21 Variables and Hypothesis ......................................................................................21 Game Theory .........................................................................................................22 Other Decision Making Models.............................................................................33 Research Design.....................................................................................................37 CHAPTER 4. JOHN F. KENNEDY: BAY OF PIGS .......................................................46 Historical Context ..................................................................................................46 John F. Kennedy and the Bay of Pigs....................................................................51 Rational Actor Model ............................................................................................61 Determining a Different Utility Function ..............................................................67 A Different Utility Function ..................................................................................70 Conclusion .............................................................................................................71 CHAPTER 5. LYNDON JOHNSON: ESCALATION OF VIETNAM ...........................73 vii Historical Context ..................................................................................................73 Lyndon Johnson and Vietnam ...............................................................................76 Rational Actor Model ............................................................................................84 Determining a Different Utility Function ..............................................................90 Governmental Politics Model ................................................................................93 Conclusion .............................................................................................................95 CHAPTER 6. RICHARD NIXON: WATERGATE .........................................................96 Historical Context ..................................................................................................96 Richard Nixon and Watergate................................................................................99 Rational Actor Model ..........................................................................................109 Determining a Different Utility Function ............................................................120 Elitist Actor Model ..............................................................................................123 Conclusion ...........................................................................................................127 CHAPTER 7. WILLIAM JEFFERSON (BILL) CLINTON: MONICA LEWINSKY SCANDAL.......................................................................................................................129 Historical Context ................................................................................................130 Rational Actor Model ..........................................................................................136 Determining a Different Utility Function ............................................................142 Governmental Politics Model ..............................................................................144 Conclusion ...........................................................................................................146 CHAPTER 8. ANALYSIS, SUMMARY AND CONCLUSION ...................................147 Kennedy and the Bay of Pigs...............................................................................150 viii Johnson and the Escalation of the Vietnam War .................................................151 Nixon and Watergate ...........................................................................................152 Clinton and the Lewinsky Affair .........................................................................154 Rational Actor Model ..........................................................................................155 Conclusion ...........................................................................................................160 REFERENCES ................................................................................................................162 APPENDIX A..................................................................................................................168 ix LIST OF TABLES Table 2.1. Typical Prisoner?s Dilemma Game Theory Matrix .........................................12 Table 2.2. Typical Free Rider Game Theory Matrix ........................................................13 Table 3.1. Two Person Game............................................................................................22 Table 3.2. Public Player....................................................................................................28 Table 3.3. Private Player...................................................................................................29 Table 3.4. Homeless Player ..............................................................................................29 Table 3.5. Nuclear Build Up.............................................................................................32 Table 3.6. Major Presidential Mistakes (1950-2000) .......................................................39 Table 3.7. Presidential Mistakes Making the McConnellCenter?s Top Ten List and Occurring between 1961 and 2000 ....................................................................................40 Table 3.8. Key Elements of the Decision-making Models Applied to the Game Theory Process ...............................................................................................................................44 Table 4.1. Bay of Pigs Game Theory Matrix....................................................................62 Table 4.2. Payoff of Different Options for Kennedy Concerning the Bay of Pigs Operation............................................................................................................................64 Table 4.3. Payoff of Different Options for Cuban People Concerning the Bay of Pigs Operation............................................................................................................................65 Table 4.4. RAM Prioritization for Kennedy-Cuban People .............................................66 Table 4.5. Bay of Pigs Game Theory Matrix....................................................................67 Table 4.6. Bay of Pigs Game Theory Matrix under Kennedy Utility Function................71 Table 5.1. Johnson-Minh Game Theory Matrix ...............................................................85 x Table 5.2. Payoff of Different Options for Johnson over Vietnam ..................................87 Table 5.3. RAM Prioritization for Johnson-Minh ............................................................89 Table 5.4. Johnson-Minh Game Theory Matrix under RAM ...........................................89 Table 5.5. Johnson-Minh Game Theory Matrix under GOM...........................................95 Table 6.1. Nixon Game Theory Matrix ..........................................................................117 Table 6.2. Payoff of Different Options for Nixon over Watergate.................................118 Table 6.3. RAM Prioritization for Nixon-Other Players ................................................119 Table 6.4. Nixon-Other Player Game Theory Matrix under RAM.................................120 Table 6.5. EAM Prioritization for Nixon-Other Players.................................................126 Table 6.6. Nixon-Other Player Game Theory Matrix under EAM with All Alternatives126 Table 6.7. Nixon-Other Player Game Theory Matrix under EAM.................................127 Table 7.1. Jones Deposition Game Theory Matrix.........................................................138 Table 7.2. Payoff of Different Options for Clinton Concerning the Jones Deposition ..139 Table 7.3. Payoff of Different Options for the Other Players Concerning the Jones Deposition........................................................................................................................140 Table 7.5. RAM Prioritization for Jones Deposition Game............................................141 Table 7.5 Jones Deposition Game Theory Matrix..................................................142, 145 Table 8.1. Presidential Mistakes Making the McConnell Center?s Top Ten List and Occurring between 1961 and 2000 ..................................................................................148 Table 8.2. Key Elements of the Decision-making Models Applied to the Game Theory Process .............................................................................................................................149 Table 8.3. Summary of Presidential Mistake Cases Studied .........................................155 xi 1 CHAPTER 1 INTRODUCTION Predicting human behavior has been the holy grail of many disciplines such as psychology, economics, sociology, and political science. When trying to predict human behavior, one does not ask why someone did something but what someone will do. To be able to predict what someone will do could make the difference between getting rich or going bankrupt or between winning a war or paying reparations. Even a small edge in predicting what an opponent will do could make a big difference. Opponents or players do not make decisions in a vacuum. They are influenced by what other players do and what they want to win. If someone is selling a house, there are only two variables involved: the market value of the house and how quickly the owner wants/needs to sell it. For the buyer, the variables are much more numerous. They may include the location, price, design features, size, and local school. Finding the key variable, the one that is most important to the buyer is difficult to do. That is what this study is all about ? determining how to determine what primarily motivates someone. If that variable can be approximated, then decisions can more easily be predicted. Behavior that must reach an accommodation with others is what is of interest here. Deciding what to wear or eat boils down to taste and availability and the day?s planned activities, but any decision that involves other people requires that one engage in 2 the dance of life called bargaining. This is the give and take between individuals, families, companies and nations. How one goes about performing the dance is their strategy. Someone using a ?hard ball? strategy will refuse to find an accommodation with his opponent but will demand that his opponent give up. (Shelling 1960, 5) One way to characterize a strategy is by using game theory. With game theory decisions made by a player are based on what he or she thinks the other player will do. A potential criminal may decide not to commit a crime if he thinks the police will catch him and put him in jail. The crux of game theory is that all players are rational human beings that will balance risks and benefits resulting in a decision that is consistent from one player to the next if provided with the same risks and payoffs. The decision making process should not be inherently different when an individual of great political power makes choices than when choices are made by the common man or woman. Individuals continually make numerous choices between alternatives, but these choices are not placed under scrutiny unless they significantly affect other people either positively or negatively. Choices that result in negative consequences are usually made by criminals, famous public figures such as actors, and athletes or politicians. These choices are put under intense scrutiny by the press and sometimes there is enough background available to determine who is involved, what alternatives were available, and the context of the situation. This is enough information to analyze the situation under game theory. Under game theory, a rational actor or player would choose an alternative which offers the lowest possibility of disaster for the greatest payoff. Our entire justice system 3 is based on the rational actor model?we hope that the consequences of going to jail will prevent most people from selecting the ?rob a bank? alternative when deciding how to earn a living. So if presidents are rational actors, should not their choices follow this model? When presidential choices, especially choices which resulted in negative consequences, are analyzed using game theory, game theory is not able to predict which choice the president would make. But if one was to use the same framework used in game theory to set up the decision making situation but apply different rules, rules derived from other individual decision making theories, then perhaps the decisions would make more sense. There are a variety of individual decision making models. The best known include the rational actor model, upon which game theory is based, the political actor model, the organizational actor model, the elitist actor model, and the idiosyncratic actor model. The political actor model bases decisions on the results of bargains and compromises of many various and diverse political players with the goal of retaining as much political power as possible. The organizational actor model bases decision making on the culture, interests and operating procedures of the organization with the goal of protecting and growing the organization. The elitist model is similar to the political actor model where the goal is to retain and increase power, but there is much less bargaining and compromise. Finally, the idiosyncratic actor model bases decision-making on personality traits such as integrity, aggressiveness, propensity to take risks, and leadership style. It is primarily a psychological approach to explaining decision-making. None of the models makes predictions, except the rational actor model using game theory; the others only provide explanations for decisions already made. But if one 4 considers that individuals with political power make decisions differently from normal individuals, then perhaps we can still use game theory to analyze the situation. Once a situation is broken down using the game theory process by alternatives, consequences, and rules of the game, the one might be able to explain the outcome of a political decision by applying the goals of the political, organizational, or elitist actor to explain or even predict a decision. This is what this study will try to accomplish. 5 CHAPTER 2 REVIEW OF THE LITERATURE: SELF-INTEREST, THE RATIONAL ACTOR MODEL, AND GAME THEORY Self-interest has always been a very compelling way to explain human behavior. Why would anyone freely do anything that did not benefit himself or herself in some way? Yet, self-interested people can be motivated by many things. Abraham Maslow, known for his hierarchy of needs (physiological, safety, love/belonging, status (esteem), and actualization), speculated that humans must attend to the lower needs before trying to satisfy the higher needs (Maslow, 1943). So, unless one knows what need or needs someone is trying to satisfy, self-interest alone can not explain or predict behavior. For instance, one cannot predict how legislators will vote on a bill to increase prison space unless one also knows whether they perceive their constituency?s desire for safety to be higher than their desire for lower taxes and therefore a higher standard of living. The model used to represent self-interest is the rational actor model (RAM). The RAM began when the philosopher Thomas Hobbes declared that people were self- centered. It was further refined by the economist Adam Smith as an explanation for how the market could provide for collective welfare when the primary motivation of individuals was self-interest. A person who is motivated by self-interest is called a ?rational actor.? The individual actor can be a person, a company, an interest group, or 6 any other political organization. Smith said that, in a free market, buyers and sellers pursue their own interests, resulting in fair prices and a healthy market. (Monroe 1991, 1) Kenneth Arrow defines a rational actor as an individual who takes actions which promote his or her preferences. Arrow does not attempt to determine what specific needs or values a person is trying to satisfy, which he and other economists term a ?utility function,? because these needs or values are unknowable without being able to read a person?s mind (Arrow 1951, 3). Arrow uses a prioritized list of preferences to determine what outcomes a person (or rational actor) would prefer (Arrow 1951, 9). An actor looks at all available alternatives and prioritizes them from most to least desirable with some being equally desirable and some being of no interest to the actor or as being inappropriate in his or her current context (Arrow 1951, 12). This prioritizing of preferences that are of interest and apply to the situation represents the values of the rational actor in a simple straightforward way (Arrow 1951, 18). RAM, as described by Arrow, assumes that an actor will look at all available alternatives, discard the ones that are not practical (such as something he or she cannot physically do or cannot afford) and selects the alternative that is most desired from the ones left. If the actor can also put all the practical alternatives in a priority order so that for every two alternatives he or she has a clear consistent preference, then the alternatives are considered ?complete.? The preference is ?transitive? if it is consistent among pairs so that where alternative A is desired more than alternative B and B is desired more than alternative C, then A is also desired more than C. If alternatives are complete and transitive, then they can be placed in a prioritized order that can be used to explain the actor?s decision (Arrow 1977, 608). 7 Anthony Downs (1957) defines a rational actor in almost the same way as Arrow as one that makes decisions which contribute toward the achievement of the actor?s goals, but Downs establishes a criterion for an actor?s prioritization. According to Downs, if it is known what an actor wants to accomplish, his or her action can be predicted by picking the most direct and reasonable actions for obtaining the goal. Direct and reasonable mean that the actor will use what he or she thinks is the least amount of resources (e.g., time and money) or the most efficient way needed to ensure the accomplishment of the goal. But what one actor thinks is the most efficient way may be different from what another actor thinks is the most efficient way for the same goal. Perhaps it is more efficient to take actions that are illegal or in a gray area than to follow the law or the current process. Both are rational within each paradigm. The illegal actions are rational because they are the most efficient for the resources. The legal actions are rational because the actor does not want to risk jail or ridicule by breaking the law or skirting established procedures. Knowing the actor?s goal and what an actor thinks is the most efficient path to the goal is key to making an informed prediction as to what decisions he or she will make (Downs 1957, 4). Using the RAM, a single decision that is made to accomplish a single goal is simple to define and analyze, but defining and analyzing many, sometimes conflicting desires, can be more difficult (Downs 1957, 4). In the study of economics this is simplified by attributing to everyone in the market a single goal of maximizing profit. The goal is never viewed as rational or irrational, only the process by which the goal is obtained (Downs 1957, 5). Wanting to own swamp land in Florida is not irrational but paying much more than market value is. 8 Downs defines a rational actor as one who can always decide between alternatives, and whose rankings are complete and transitive and do not change unless the goal changes. Also, a rational actor who initially appears to act irrationally will rectify his or her mistake when he or she discovers it whereas an irrational actor will continue to repeat the same mistake over and over. Downs sees the ability to correct a mistake as more important than making the mistake when determining rationality (Downs 1957, 9). When considering rationality, Downs considers only economic and political goals. He does not take into account personal goals. Downs considers it irrational for someone who prefers a Democratic candidate to vote for a Republican candidate in order to please a spouse because his or her goal was personal rather than political (Downs 1957, 7). Using Down?s definition of rationality unrealistically places very narrow limitations on goals. These limitations may work when dealing with large groups such as voters or political parties since, in general, they follow political and economic goals, but it may not work so well when looking at individual decision making. Downs? theory of rationality is based on the assumption that government?s goal is to obtain as much political support as possible, and the goal of the political parties and politicians is to get reelected (Downs 1957, 11). James Buchanan made the leap from economics to politics saying that political relationships are similar to market relationships in that actors ?exchange? support for candidates and issues rather than goods and services and that each actor tries to maximize power rather than money (Buchanan 1962, 19-20). Stephen Holmes points out that rational behavior is considered calculating self- interested behavior. By this definition, any action someone purposefully decides to take that will provide him or her some gain can be considered rational behavior (Holmes 9 1990, 268). A suicide bombing would be considered rational behavior if it was purposely done, as evidenced by its planning, and the bomber expected to get a free pass into heaven. Holmes tries to separate actions resulting from calculated self-interest and those resulting from impulsive passion. Emotions such as hate, fear, envy, anger, and grief prevent a person from making a rational well considered evaluation of alternatives resulting in self-destructive, rather than self-interested, behavior. This way, the rational selfish person can be much more moral than the irrational selfless person (Holmes 1990, 268). Religious zealots of medieval times exacted untold tortures and other cruelties on those accused of sacrilege. These actions probably did nothing to materially benefit the torturer. In this kind of situation the leaders of a passion motivated movement often gain a benefit, such as power or wealth, while the followers are motivated by passion or survival (avoiding persecution). Also, what may have started as self-interest, such as the oppression of Africans as slaves for an economic benefit, could morph into the oppression of African Americans but with no discernable benefit. (Holmes 1990, 271- 273) Many opinions have little to do with self-interest. Those who profess to be pro life could see increased welfare spending, more crime, and higher population density in the inner cities if abortion were abolished (Levitt 2005, 136-144). It may be irrational for a voter to be against abortion, but it would not be for a politician, if it got him or her elected (Holmes 1990, 274). Where emotion brings irrational, self-destructive behavior, elites can often still benefit from supporting such behavior. History is littered with actions that defy self-interest because people are insecure, frightened, or just following 10 the latest fad. Yet floating to the top are leaders and aristocrats who rationally take advantage of these irrational actions (Holmes 1990, 275). A utility function or goal of ?peace of mind? cannot be considered rational when forcing others to adhere to a particular religious belief, but it could be considered rational if power or resources could be gained for a leader by adding new members (Holmes 1990, 278). Democracy and communism are attempts to resolve the interests of leaders with the interests of the masses. Under communism the leaders would theoretically eventually disappear and all actions would be taken in the interests of the masses. In a democracy the citizen votes for a leader who professes to have the same interests as the voter. Once in office, the leader is expected to pursue the voters? interests (Holmes 1990, 285). Under the RAM, as described by Arrow, Downs, and Buchanan, an act of charity or altruism would be difficult to explain. It would not seem to be in a person?s self- interest to purposefully decrease their resources knowing that they will get nothing in return. Peter Hammond proposes that one reason a person donates to charity may be in the hope it will encourage others to also donate so that the charity will be available if he or she ever needs it (Hammond 1975, 116). Hammond discusses a situational experiment called the ?poverty game? where one player gets food on odd days and another player on even days. The food must be eaten on the day provided. The players share the food as long as the game is continuous and both players cooperate (Hammond 1975, 117). So they provide charity when rich only with the expectation of receiving charity when poor. This theory can be extrapolated from one day to fifty years or so where a person gives money to the poor when young with the expectation that he or she will receive money when old and poor (Hammond 1975, 119). Hammond does not address the issue of 11 collective action where individuals do not cooperate when they think that they would benefit even if they never contributed. This is often referred to as the ?free rider? problem and is more consistent with the RAM. The RAM does not generally delve into how an actor determines what preference will provide the greater utility. It is game theory which takes the next step of presenting all alternatives in an orderly manner and determining which outcome provides the greatest utility. (Monroe 1991, 4) Game theory uses a RAM which predicts that a rational actor or player, when presented with a situation requiring a decision that is affected by other players, will choose an alternative that offers the lowest possibility of disaster for the greatest payoff (Brams 2004, 30). Brian Barry and Russell Hardin view the free rider problem as similar to a prisoner?s dilemma game under game theory (Brian and Hardin 1982, 24). In the prisoner?s dilemma game, two criminals are being interrogated by the police and cannot communicate with each other. They each have two choices, they can tell the police everything they know about the crime (defect to the police) or they can be silent (cooperate with each other). The result or outcome depends on what the other criminal chooses. If both criminals have two choices, then there are four possible outcomes. If they both cooperate by not talking to the police, they will each receive the same small pay off (i.e., one point) since the police will not have enough evidence to prosecute them for all their crimes. If they both talk to the police (defect), they will get no pay off since the benefit of coming forward will be balanced out by the police?s full knowledge of their crimes. If only one player defects and the other cooperates, then the one who defects will get a higher payoff (i.e., three points) than the one who cooperates (i.e., negative one 12 point) since the defector will get a lower sentence for his or her testimony (Brian and Hardin 1982, 24). A graphical representation is presented in Table 2.1 below. Table 2.1: Typical Prisoner?s Dilemma Game Theory Matrix Player 1 Matrix 1 Cooperate Defect Cooperate 1, 1 -1, 3 Player 2 Defect 3, -1 0, 0 Source: Brian and Hardin 1982, 24. This means that an individual player can get his best score of three by defecting. If both players cooperate, they can do better than if they both defect, but by cooperating they risk their worst pay off of negative one. So each player?s dominant, or best strategy, according to game theory, would be to defect (Brian and Hardin 1982, 24). Game theory assumes that rational players will pick the alternative that gives them the biggest payoff they can win, assuming their opponent is equally rational and selects the alternative which will give the other player the smallest maximum payoff (Davis 1983, 22). Assuming player 1 defects, the best player 2 can do is to defect also. According to Brian and Hardin, the prisoner?s dilemma graphically represents the free rider problem, as shown in Table 2.2 below. 13 Table 2.2: Typical Free Rider Game Theory Matrix Citizen 1 Matrix 2 Clean Up Do Nothing Clean Up 1, 1 -1, 3 Citizen 2 No Nothing 3, -1 0, 0 Source: Brian and Hardin 1982, 25. Suppose citizens 1 and 2 have a common driveway that needs to be cleaned up. If neither cleans up the driveway, there is no pay off since the driveway remains the same. If they both help, there is a pay off of 1, since they both now have a clean driveway. If one does all the work and the other does nothing, then the player doing nothing gains the pay off of a clean driveway but with none of the work (Brian and Hardin 1982, 25). So, like the prison?s dilemma, the rational decision is to do nothing since citizen 2 will assume citizen 1 will do nothing also (citizen 2?s worst outcome). So, how can the rational actor theory be supported in the face of noncompulsory actions that support the common good but do not maximize individual objectives? Either the theory is flawed or the players are getting other benefits from participating in collective actions such as voting or joining interest groups. The ?by-product? theory speculates that people join groups for more reasons than the hope that the group will attain its goal. Some may join for low cost group insurance, for the publications produced by the group, or for social contacts (Olson 1965). The problem with this theory is that many groups, such as environmental groups like the Sierra Club, offer very little extra motivation to join and yet they continue to attract members. Also, it does not show how or why groups initially form (Brian and Hardin 1982, 26). 14 Game theory itself gives some clue to the free rider problem. When prisoner?s dilemma is played only once, then defection would be the most rational outcome. But when the same game is played numerous times, cooperation becomes more rational because if one player always defects, then so will the other and the pay off at some point in the super game (series of many games) will be zero. But if the players decide to trust each other and cooperate, like in the poverty game described earlier, they can increase their pay offs to one for every game played. This would be rational only if the number of players is small enough for other players to notice and respond to a defection. If the number of players gets too large, then the rational action would again be to defect (Brian and Hardin 1982, 34). Although the free rider problem has been difficult for the RAM to address, the more difficult problem is to determine each actor?s goal or utility function. Buchanan uses the term ?economic model of behavior? to refer to the RAM. In his theory, Buchanan specifies that the primary motivation of the individual is ?utility maximization.? Once it can be established what the individual wants to accomplish, then the theory can be tested (Buchanan 1972, 16). Buchanan sees this theory as predictive of what the individual will do and makes no value judgments as to whether the actions taken are good or bad (Buchanan 1972, 17). Jane Mansbridge (1990) points out that what can comprise a utility function under the RAM can vary considerably from theorist to theorist. On one end of the spectrum some think that only economic self-interest should be considered while others will include status or power. On the other end of the spectrum, everything is fair game, including motives that are clearly self-destructive and irrational (Mansbridge 1990, 256). 15 Most theorists consider the self-interest of an actor?s family and friends to be part and parcel of the actor?s self-interest, but this expansion of self-interest is difficult to define. In some contexts one might also include members of the actor?s race or religion or even all mankind (Mansbridge 1990, 257). Such an expansion of the definition of self-interest would certainly account for altruism. An actor may be pulled in opposite directions between two utilities such as the desire to be reelected and the desire to follow one?s own principles (Mansbridge 1990, 258). As an actor?s wealth and power increase, the incidences of following one?s principles will often increase as well. If one already feels secure, then self-interest may take a lesser role to charity, morals and principles (Mansbridge 1990, 259). This harkens back to Maslow?s hierarchy of needs?once the lower needs are met, one is free to pursue the higher needs. Federal judges are appointed for life based on the conviction that a judge who no longer has to worry about job security will be relatively unlikely to be corrupted. A president in his second term is no longer interested in reelection but may want to secure a legacy based on sound moral principles. So perhaps a president in his second term would be more likely to base decisions on his own moral principles or what he thinks is best for the country, unless the decision involves avoiding impeachment (Mansbridge 1990, 260). The tendencies described above are generalities based heavily on context. Actor personality, social norms and organizational roles and culture have a particular bearing on an actor?s choice between self-interest and principles. For instance, a senator on a general committee (such as Veteran?s Affairs) may vote for the overall good of the country more often than when on a specific committee that would directly support his or 16 her state (such as agriculture). A marine who is trained to protect his team mates (organizational role and culture) may put their self-interest over his or her own on the battlefield (Mansbridge 1990, 261). Although putting one?s life on the line for one?s comrades in arms can be compared to a super game where the marine provides protection in one game with the expectation that he or she will get protection in the next, understanding an actor?s personality and the context in which the decision is made is important to understanding which utility function will dominate the situation (Mansbridge 1990, 261). Broadly stated, the purpose of this study is to find an objective method to determine the most likely dominate utility function that will predict political actions. Dennis Mueller proposes two steps in constructing a valid rational actor model (RAM) for testing. First of all, the definition of the utility function must come from other disciplines than economics. If we are looking at any kind of human behavior, we can not necessarily assume that maximizers (those seeking the maximum self-interest) only seek money and power. The second step is to determine what constraints the actor is under. Monetary constraints are normal when doing an economic analysis, but, here again, other constraints, such as religious or ideological beliefs, may come into play and must be considered. (Mueller 2003, 659) During the mid 1990s the RAM branched off into the subspecialties of public choice, rational choice and social choice. These subspecialties are very similar and the names applied to each are often used interchangeably. Public choice looks at how decisions are made using political mechanisms rather than market mechanisms. Rational choice looks at the decision making process rather than the actions taken, and social 17 choice looks at how voting and justice systems can support social welfare. (Monroe 1991, 2) While discussed by many theorists, the RAM continued to basically represent purposeful, goal directed self-interest as described by Holmes and others many years ago. According to Kristen Monroe there are seven key assumptions associated with the RAM. These are as follows: 1. Behavior is goal directed 2. Goals are perceived to be in the actor?s self-interest 3. The actor will make a conscious choice between alternatives 4. The basic unit of measurement is the individual 5. The actor can rank outcomes in a priority order that will not easily change 6. The outcome with the higher priority will provide the most utility 7. Actors have enough information to make a determination (Monroe 1991, 4) The RAM has been criticized for several reasons over and above the free rider problem. Some theorists think it is too limited since it does not factor in cultural norms. Others criticize the theory as too general where anything an actor does could be considered rational if only his or her real goal was known. (Monroe 1991, 6) Along with free riding, voting is most often used to discredit the RAM. Under a RAM based solely on maximizing power and money, no one would ever vote since the effort of finding out about the candidates and going to the polls does not justify the impact of one vote. Yet about fifty percent of eligible voters do vote. Does that mean RAM is half correct or does it mean that more factors than economic self-interest are 18 coming into play here? The same can be said for the free rider situation where only some, but not all people, contribute to the public good. (Mueller 2003, 267) Why do some contribute and not others? Should voters and those who contribute to a public good be considered irrational? If so, is there a difference between the rational and the irrational actors where the rational actors are perhaps goal driven and the irrational actors are emotion driven (Holmes 1990, 268), or do we just not understand the utility functions involved? A third criticism of RAM is based on Herbert Simon?s ?bounded rationality? which says that individual decisions are not based on getting the most utility but on ?satisficing? where the individual is intent on gaining only a minimal amount of ?satisfaction? and once that minimal level is obtained, does not go out of his or her way to seek more. (Monroe 1991, 7) The last major criticism is that the prioritization of preferences by an actor is not consistent over time, nor is an actor?s analysis of the information readily available. An actor?s priorities may change depending on context. An actor may have many roles such as parent, professional, child, church member, and/or environmentalist. Each role comes with unique goals and may become dominant based on the context of the decision. (Monroe 1991, 21) It is clear from this discussion of the RAM that political decision making cannot be based on economic and/or reelection self-interest alone. Humans are complex creatures and many other factors come into play such as personality, culture, norms, context, roles, perception of other player?s strategy, and experience. The question here is 19 whether this list of factors can be simplified in order to reliably predict what actions an actor will take. One possibility is to consider decision making models other than the RAM. The most widely used, other than the RAM, are the governmental politics model (GPM), the organizational behavior model (OBM), the elitist actor model (EAM), and the idiosyncratic actor model, although names vary from theorist to theorist. The GPM bases decisions on the results of bargains and compromises of many various and diverse political players with the goal of retaining as much political power as possible (Truman 1971, 501-536; Allison and Zelihow 1999, 255-313; Neustadt 1991, 47-49). The OBM bases decision making on the culture, interests, and operating procedures of the organization with the goal of protecting and growing the organization (Allison and Zelihow 1999, 143-185; Halperin 1974, 26-62). The EAM is similar to the political actor model where the goal is to retain and increase power, but there is much less bargaining and compromise because the central actor holds more power than anyone else (Dye 2001, 1-15; Jones 1994, 8; Mills 1956, 3-29). Finally, the idiosyncratic actor model bases decision-making on such personality traits as belief system, character, style, and worldview (George and George 1998, 152-157; Barber 1985, 1-11). It is primarily a psychological approach to explaining decision-making based on individual differences rather than the generalization that are usually used by creators of decision making models. Books based on the idiosyncratic actor model will be used in this study only to justify the choice of decision making model applied to each president. None of the models makes predictions, except the RAM using game theory. The others only provide explanations for decisions already made (Austen-Smith and Banks 20 1998, 259). But by using the RAM alone, one could never have predicted that Bill Clinton would have made the choices that resulted in his impeachment or that George H. W. Bush would have raised taxes after promising not to. The Present Study Given that individuals with political power do not always seem to make rational decisions, then perhaps we must apply different models (Mueller 2003, 659) in determining the best utility function, so we can continue to use game theory to better analyze these seemingly non-rational decisions. Once a situation is characterized using the game theory process by alternatives, consequences, and rules of the game, one might be able to predict the outcome of a political decision by applying the goals of the political, organizational, or elite actor (Ordeshook 1986, 2). The individual actor analyzed will be a past president of the United States. Each decision examined will be one widely considered to be a blunder. Each decision or case will be analyzed with RAM only and then by RAM using three well known political/sociological/psychological models to better home in on a more precise utility function then power and money. This will all be operationalized in more detail in the next chapter. Gabriel Almond points out that there has been little effort to integrate sociology and psychology into the RAM, and this has been a criticism of the theory for over fifteen years (Almond 1991, 48). This study will contribute to the rectification of that criticism. The next chapter will lay out the research design, provide an in-depth discussion of game theory and other decision-making models, and present a justification for the case studies selected. 21 CHAPTER 3 METHODOLOGY The purpose of this chapter is to describe the research design, provide an in-depth discussion of game theory and other decision-making models, and lay out a justification for the case studies selected. Research Question The research question stems from Gabriel Almond?s (Almond 1991, 48) astute criticism of the rational actor model (RAM). Can sociology and psychology be integrated into the RAM without significantly changing the nature of the model, and if so, does this integration improve the model to the point where it can better predict political behavior? The central hypothesis of the study is that such an integration can be made where the RAM would still retain its integrity while providing a better prediction of individual political actions. Variables and Hypothesis The hypothesis of this study is that when other decision making models (such as governmental politics model (GPM), organizational behavior model (OBM), elitist actor model (EAM)) are used with the RAM, a more complete explanation of decisions can be provided than by using the RAM alone. In order to test this hypothesis, a qualitative research design must be used since quantitative designs do not collect data about the context under which actions occur. Quantitative designs collect a limited amount of 22 specific data for a large number of cases while qualitative research can be more detailed about a small number of cases (O?Sullivan and Rassel 1999, 36). This type of approach is necessary for game theory since many details must be collected on each game in order to predict how the game will be played using the RAM. These details are also needed to objectively select an appropriate model for the determination of a utility function. Game Theory Game theory is a framework for making future decisions as well as analyzing past decisions. A game consists of players who must choose between alternatives in order to reach a goal. Each alternative has different payoffs depending on which alternatives the other players choose. A single game can be set up when two or more people or groups (players) have a decision to make that affects the outcome of the other player?s decision. A simple game can be characterized using a matrix with one player depicted on the left and one above. It is assumed that they both select from the same alternatives, but the results may be different for each depending on what the other player selects. Also, the players might not have the same goals and objectives, making the payoff for an alternative different for each player. A simple matrix looks like Table 3.1 (Davis 1983). Table 3.1: Two Person Game Person B Alternative X Alternative Y Alternative X Player A payoff, Player B payoff (if both pick alternative X) Player A payoff, Player B payoff (if A picks X and B picks Y) Person A Alternative Y Player A payoff, Player B payoff (if A picks Y and B picks X) Player A payoff, Player B payoff (if both pick Y) 23 (Based on Davis 1983, 76) It is assumed that the player will choose the alternative that gives the highest payoff assuming the other player chooses an alternative that would give the worst payoff. Under classical RAM, economic payoffs are based on money while political payoffs are based on maximum gain and minimum risk. Games can take the form of a single play or repeated plays. Play can be simultaneous so one player makes his move at the same time as the other player, or moves can be consecutive where a player moves after he sees the other player?s move, as in chess. Games can also be zero sum, positive sum, or negative sum. A zero sum game means that the players have opposite goals. When one player wins, the other player loses that same amount so that the total of winnings and losses is zero. Positive sum games mean that both may win, as in the marketplace. In negative sum games (a.k.a. negative expectation) the total of the players? winnings do not equal the amount in play, such as casino games where over the long run the players lose a percentage of money to the house (Davis 1983, 14). Game theory uses a rational player model that assumes that rational players will pick the alternative that gives them the biggest payoff they can win assuming their opponent selects the alternative which will give them the smallest maximum payoff. If the biggest minimum payoff equals the smallest maximum payoff, then the game has an equilibrium point. The equilibrium point in a two person zero sum game is considered the solution because neither can do better by changing his or her choices (Davis 1983, 22). 24 When games have no equilibrium point, each player must try to predict what the other player will do. Assuming the players are equally intelligent, it is sometimes beneficial to pick alternatives at random with equal probability, making it impossible for the other player to predict the choices (Schelling 1980, 175). For instance, a general might pick targets at random so that the enemy will be taken by surprise. This is called a ?mixed strategy.? The player using a mixed strategy will do better than an opponent who uses a single or pure strategy if the strategy can be predicted (Davis 1983, 52). Under game theory, to solve a two player zero sum game (what one wins, the other loses) one would first look for an equilibrium point where each player selects the alternative that provides the best worst case (i.e., the worst thing that could happen in alternative A is a lot better than the worst thing that could happen for alternatives B or C). If the alternatives they select give them each the same result (best worst case), then the players have selected the best solution for both of them and the game is solved. But if there is no equilibrium point, then each player must eliminate all the alternatives where the best result is worse than the worst result of other alternatives. In other words, if the worst case of alternative A is better than the best case of alternative B, then B can be eliminated as an alternative. B is referred to as a dominated strategy. Using the remaining alternatives (also referred to as strategies) a player randomly picks each alternative with such a frequency that, on average, each player does as well as he or she can based on the payoffs (Dixit 1991, 61). In the world of mathematics where numbers can be assigned to alternatives and moves are made repeatedly, a game can always be solved. But in the real world, this is much more difficult to do. Usually a player does not get to play the exact same game 25 more than once. Although there can be a sequence of games with moves in each, players make one move at a time, re-evaluate their position, and then start a new game for their next move. Also, setting up a matrix of alternatives and assigning values to results based on what other players do in order to determine the best alternative is very subjective. Usually the best one can do is rank order his/her result preferences and pick the alternative that gives the best worst case (Muzzio 1994, 51-54). This strategy is sound because, if the other player plays the same way, each gets his/her best worse case. But if both players pick the alternative that gives each his/her best best case, then anything can happen. Players who do this are often considered to be impulsive risk takers and to have poor judgment. Politicians, especially those who have reached high levels, often make decisions that appear risky or non-rational; this characteristic may result in a disastrous result, either for the country, for their career, or both (Reedy 1970, 27). Every president since Eisenhower has made such decisions. This idea will be discussed later in the chapter. There is a fundamental difference between game theory when it is used to explain economic decisions than when it is used to explain political decisions. Normally in game theory the games are played by private individuals making rational decisions between alternatives based on what they think other private rational individuals might do. The market is made up of many decisions or interactions among buyers and sellers attempting to find a mutually agreeable price (Buchanan and Tullock 1965, 33). In game theory, players are trying to find strategies that result in the best outcomes. Game theory predicts that a player will not make a decision that has the possibility of the worst or most disastrous outcome, yet many political decisions are made that include the possibility for 26 disaster (Brams 2004, 5). This may be because those decisions also include the best possible outcome or perhaps the worst outcome in the mind of a private person is different from the worst outcome in the eyes of a public person. For example, President Jimmy Carter?s decision to rescue the hostages in Iran had chances for both the best and the worst outcome; so rather than continue to negotiate for the release of the hostages, Carter choose to attempt a very difficult and dangerous rescue that resulted in eight American deaths. Of course, if President Carter?s attempt to rescue the hostages had been successful, it would not have been considered a mistake. It is the failure that puts it under so much scrutiny. Under the RAM, a decision is a mistake if it includes the possibility of the worst outcome not because it did not achieve the goal. How a player views the possible outcomes under game theory is called the utility function, or the attitude a player has toward risk. For instance, winning $10 is more important to a child than to an adult. A player who is wealthy is more inclined to risk $50,000 in order to win $1 million than a player who is poor. This is why it is important to understand a player?s goals and preferences and why the sociological and psychological models are important. The sociological theory similar to game theory is interactionism. Interactionist theories look at the process by which people interact with each other and determine what people should to do to increase the likelihood that their interactions will result in positive outcomes (Goffman 1981, 5-77). It is difficult to determine the correct utility function of each alternative for a player because the player?s preferences will not only change over time, but may change from moment to moment as possible alternate outcomes are re-evaluated. Also, they will change from game to game in a series of games. For example, with Bill Clinton, the first 27 game was deciding to have a sexual relationship with Monica Lewinsky, then the next game was determining how much he would cooperate with investigators, and in a later game he decided to lie under oath. Although this study only analyzes one game for each President, it is important to understand that with each game, motivations, perceptions, and preferences shift. The problem of determining utility is very similar to the problem of getting an accurate response to a survey. The wording of the questions, the order of the questions, as well as how the questions are asked (written, phone, in person) will affect survey results (Davis 1983, 63). Even if utility could be accurately measured, it seems to change depending on whether the player is risking his or her own money or someone else?s, whether someone is watching, and successes and failures in past games. For instance, losers accept greater risks than winners because they want to recoup their losses while winners want to conserve their winnings. (This does not necessarily apply to professional poker players who are risk seekers, not risk averse.) Also, players are not always forthcoming with how important a payoff is to them since they might gain more if they appeared to care less. For instance, the buyer of a used car will get the car at a lower price if he can convince the seller that he is less interested than he really is (Davis 1983, 63). The beauty of game theory is that it sets up a decision in a very straightforward and rigorous way. The problem is determining the utility function since that payoff is in the eyes of the player. To set up the payoffs, one must take into account penalties. Say someone has committed a crime and is deciding whether to keep quiet or admit the crime and provide information on other players in order to get the payoff of immunity or a lower sentence. Penalties for cooperating with the police may include loss of job, family, 28 and status or loss of membership in a mafia or political circle that promises rewards for keeping silent and penalties for confiding in the police. Twenty years in jail may seem a better result then being killed by the mafia (Muzzio 1982, 14). Factoring in all possible outcomes will dramatically change the payoff matrix. The matrix of a public individual such as a politician may look like Table 3.2. The other player will be considered a private individual to make the difference between the two players the most apparent. The numbers are not a literal number of years. They are used to show that the personal situation of the player would affect how he or she views the payoffs of each alternative. Table 3.2: Public Player Other prisoner Cooperate with player (hide truth) Cooperate with Police Hide truth 100, 10 -100, 0 Public Individual Cooperate with Police -30, -20 -50, -5 (Based on Davis 1983, 57-74) Using this matrix, the rational decision for the public individual would be to cooperate with police since he would avoid the worst case of a -100 payoff (assuming that cooperation does not result in something like a mafia hit) (Davis 1983, 57-74). The matrix of a private individual with a good job and a family may look like Table 3.3. Here, the rational decision again would be to cooperate with police to avoid the highest negative payoff. 29 Table 3.3: Private Player Other prisoner Cooperate with player (hide truth) Cooperate with Police Hide truth 50, 10 -50, 0 Private Individual with recourses Cooperate with Police -20, -20 -30, -5 (Based on Davis 1983, 57-74) The matrix of a homeless individual may look like Table 3.4 since he would only be worried about the number of years in jail. In this case the negative numbers do equate to sentence years. Table 3.4: Homeless Player Other prisoner Cooperate with player (hide truth) Cooperate with Police Hide truth 10, 10 -20, 0 Homeless Individual Cooperate with Police 0, -20 -5, -5 (Based on Davis 1983, 57-84) The pay off and penalties are highly dependent on the context. When the penalties are redetermined based on what each individual stands to lose, the game may still end the same way and not explain why some public figures choose the worst case scenario in the hope of getting the best case results. How can this be resolved? The theory of this dissertation is that only by applying other decision making models in conjunction with game theory can one explain and predict some presidential decisions (Davis 1983, 57-84). By choosing the best worst case alternative, players cut their losses, which is the theory behind making deals with the police in criminal cases or settling in civil cases 30 because going to court may result in the worst case scenario. But it may also result in the best case scenario. By cutting one?s losses, the player also loses any chance to win the jackpot. Amateur gamblers may be familiar with odds but few play it safe; most go for the jackpot which is why states make so much money on lottery tickets (Davis 1983, 82- 84). A rational player will assume that his opponent is at least as smart as he is, but is this also the case with politicians? Perhaps someone who has attained a high level of success will attribute it to superior intelligence. People who are overly confident tend to underestimate their opponents. They may also come to believe that they are invulnerable, especially after they win an election. One sometimes sees this happen with poker players who win a series of pots and appear to come to the belief that they can not lose. The opposite also occurs. A player who suffers a bad beat (losing a large pot with a strong hand to a player with an even stronger hand) may begin to play wildly and irrationally and usually loses still more (Davis 1983, 82-84). In the two person non-zero sum game, it is possible for both players to win or both to lose. Players who cooperate could possibly win more than players who play against each other. This type of game tends to represent more realistic situations. Non- zero sum games could require total cooperation for the players to win, minimal cooperation, or anything in between. The main point of non-zero sum games is that the players have three alternatives: 1) Play totally for themselves and end up with nothing or worse; 2) Play conservatively and win at some minimal level; 3) Cooperate and maximize their winnings; 31 (Davis 1983, 82-84) Positive sum games are more complex than zero sum games because communication between players can have a significant effect on the results. In the zero sum game communication should have no impact on the results, since what one side wins the other side loses, but in the positive sum game players must understand the rules of the game and the rules about communications among players along with who plays first, how much information each player has, whether agreements between players are binding and how many times the game will be played (Davis 1983, 82-84). In a marketplace no communication between players is needed beyond the signals that the market provides. The key is for players to set up the rules so that there is a free market as in, for example, international trade with no barriers (Davis 1983, 108-112). The prisoner?s dilemma is the classic non-zero sum game played where players are not allowed to communicate but cooperation between players who committed a crime together would give them the shortest jail term. Yet cooperation with the other player could also result in the worst outcome (Davis 1983, 108-112). Games such as this require cooperation for both players to come out ahead. Noncooperation would provide one player a huge benefit only if the game is played once. But for games played repeatedly with the same players, noncooperation could result in disaster for both players. Whether players cooperate or not depends on context, player personality, player understanding of the game, the amount of penalty/reward, whether players can communicate, and if agreements are binding (Davis 1983, 157-160). The nuclear deterrence of the Cold War was a two player non-zero sum game that looked something like Table 3.5, where the numbers represent billions of dollars saved 32 and increased national security. It is obvious that there is much benefit for both sides if they cooperate and reduce their nuclear capabilities. They would both save money and decrease the chance of a nuclear war. By the same token, both players increasing their nuclear capabilities would cost billions and would threaten the safety of the world. It cannot be determined who are the winners and losers if one player builds up and the other one reduces unless you know whether the player building up is doing it for defensive or offensive purposes (Brams 2004, 40). Even if a player never intends to use the capability, the game changes if nuclear weapons are used as a threat to force the other player to comply (Dixit and Nalebuff 1991, 217-220). This game changes again when the second player is a terrorist group rather than the Soviet Union. Under such a game the US alternatives are no longer to build up/reduce nuclear capability but to prioritize a whole set of new alternatives such as to increase intelligence capability, invade or sanction the country where the terrorist group has its base, and/or bomb terrorist camps. Table 3.5: Nuclear Build Up Soviet Union Build up nuclear capability Reduce nuclear capability Build up nuclear capability -100, -100 ? USA Reduce nuclear capability ? 100, 100 (Based on Brams 2004, 40) Then there are games of more than two players called N person games. Many political situations are N person games such as an election where the voters are the players and electing the desired candidate is the payoff. In two person non-zero sum games, cooperation is needed for both players to realize a decent payoff. In N player 33 games, cooperation is essential (Davis 1983, 167-169). A single voter can not accomplish much, but a coalition of voters can. The rules of the N person game, such as how voting occurs, are also critical. Who is allowed to vote and under what circumstances? Does the winner take all the electoral votes of a state or are they proportionally allocated? Is there a run off if no one receives a majority? Rules can make a big difference in how payers make their choices. An N person game can also be viewed as a two person zero sum game. For example, one player is all the voters in a coalition or all the politicians in a party, and the other player is everyone else. This game is purely competitive since what one side wins; the other side loses, as in a zero sum game (Davis 1983, 169). Game theory relates to many disciplines: management, economics, psychology, and political science, to name a few. Political scientists have used game theory for studying international relations, voting, vote-trading, voting power, forming coalitions, and elections (Brams, 2004, p. xvii). Using the process for setting up a game is a good way to view decision making and to analyze any decision, but the final result does not always predict or explain the real life result because game theory uses a classical rational actor model. Since there are several other decision making models, using a different model to determine the utility function may yield better explanations for past political decisions, especially for those that resulted in the worst case scenario that the rational player would never have selected under the classical RAM. Other Decision Making Models Graham Allison and Philip Zelikow, in their book Essence of Decision: Explaining the Cuban Missile Crisis (1999), use three models to explain how nations 34 make decisions. These models are the rational actor model (RAM) which is similar to classical game theory, the organizational behavior model (OBM), and the governmental politics model (GPM). Although these models are applied to nations in Allison and Zelikow?s book, they could be adapted for use in modeling the decision making process of political figures. The rational actor or classical model, according to Allison and Zelihow, is based on the assumption that a nation behaves as a single actor that has logically and deliberately looked at all alternatives and has selected the one that provides the best gain with the least risk based on a consistent set of national values and what they think other players will do. In this model, the nation makes decisions as if it were a rational individual with goals that do not easily change and can be prioritized. For instance, a country?s top goal may be domestic security while an individual?s top goal would be keeping himself and his family safe. Next the nation or individual must understand what alternatives are available to meet its goals and be able to envision the possible outcomes of each alternative based on what other players do. Finally, the alternative selected is the one with the greatest payoff and lowest risk. Unfortunately, as pointed out by Herbert Simon, individuals do not have total knowledge of all alternatives, all possible consequences, and the likelihood of each. This limitation is referred to as ?bounded rationality.? (Simon 1985, 294) Therefore, when explaining or predicting decisions, one needs to set up the analysis with what the nation or individual believed at the time about other players, not what was known later when more facts became available. The organizational behavior model (OBM) is Allison and Zelihow?s (1999, 255- 313) second model for national decision making. This theory assumes that one person or 35 even one organization does not make decisions for a nation. In the United States decisions are made by the Office of the President, Congress, and relevant organizations such as the Department of State, Department of Defense, and Central Intelligence Agency. All organizations have procedures that they follow to take the actions needed to accomplish their mission. How they are structured, their history, funding, culture, and how they interact or do not interact with other organizations all influence the prioritization of goals, the selection of alternatives, the evaluation of possible outcomes, and the final choices. Allison and Zelihow point out that this decision process is not necessarily rational. Organizations are constricted by internal goals, rules and agreements, as well as organizational culture in their decision making process. There is also a dispersion of responsibility and accountability, making it difficult to determine who should make the decision (Allison and Zelihow 1999, 143-157). Another model related to OBM is the ?garbage can? model which claims that organizations function under a form of ?organized anarchy.? This means that goals, processes, and players are not well defined and that organizations make decisions that are sometimes unrelated to problems (Olsen 2001, 191-198). This theory will be factored in to the analysis if a president appears to be operating under OBM and the ?garbage can? model also applies. As opposed to the RAM, where a nation tries to maximize its advantage and minimize risk, the OBM will do the same thing using a greatly constrained set of alternatives. These alternatives are limited to what organizations think is appropriate to do based on mission, funding, capabilities, resources, structure, and culture. What has 36 been done in the past is what is perceived to be the correct action for future similar situations. Organizations are also constrained by other organizations. It is difficult for organizations to work together cooperatively since they fear treading on each other?s turf and making themselves obsolete. This further limits alternatives (Halprerin 1974, 26-62). Using an OBM, alternatives are limited to organizational capabilities, and choices are made as individual members of an organization follow standard operating procedures (SOPs), making small choices, unaware of the big picture, which inexorably brings the organization to a result even though the organization as a whole never made a specific choice (Allison and Zelihow 1999, 152). People do this, too. If a person has an SOP of lying to cover small transgressions, each small lie will compound the problem until recovery becomes hopeless unless another strategy is chosen, such as confessing and asking forgiveness. To look at a nation as a group of organizational actors is to say that the nation?s actions are the result of the output of a variety of governmental organizations. To look at the decision making of a president from an organizational model is to say that his decisions are the result of actions and inputs from a variety of groups such as his staff, government agencies, Congress, his party, and interest groups. Allison and Zelihow?s third model is the governmental politics model (GPM). In this model decisions are made through bargaining and compromise. Goals and objectives are not consistent but depend on the strength of specific interest groups, public opinion, and constituency ideology. The rules of the game are the political rules first and foremost, which means that fund raising and keeping constituents happy are keys to staying in power. (Allison and Zelihow 1999, 255-313) 37 A model that Allison and Zelihow do not use is the elite actor model (EAM). The most well known authors of this model are Thomas Dye and C. Wright Mills. Dye describes a decision making model where large foundations established by corporate directors and other wealthy individuals fund think tanks to identify problems and devise solutions that politicians then support. Money is the source of this elite power, and the motivation is to protect the source of the money and to retain power. If you control a large corporation, you have the money and ability to influence policy (Dye 2001, 41). Mills writes that the interlinking of three major institutions?the military, corporations and political parties?has placed enormous power in the decision makers of these institutions, and these decision makers have become the primary source of power in America (Mills, 1956). A person functioning under this model will take whatever action is necessary to maintain power. Research Design Using game theory we can take a problem and predict, after the fact, what decision should have been made by a rational actor. Then we can use Allison and Zelihow?s three decision making models plus the elitist actor model (EAM) to determine if the results can be better explained with the help of these models. The problem with applying game theory is determining how a player reads other players and how he or she evaluates payoffs. The other models can perhaps shed light on these issues. The qualitative design of case study will be used for this research. Case studies look at the details surrounding and involving an incident or person, taking context into account. Case studies are the recommended design if the researcher needs to investigate how and why specific decisions are made (O?Sullivan and Rassel 1999, 38). Case studies 38 are particularly useful for the study of incidents that occur in the every day world outside of a laboratory since the event or incident itself can no longer be influenced by the researcher because it has already occurred (Yin 2003, 1-18). Extreme cases, such as presidential mistakes, are often selected for case study research because the phenomenon under study will be more apparent and therefore easier to identify and study. Case studies can not easily support a general theory of behavior due to the small sample size used, but they are particularly good at disproving a theory or rallying interest in a theory in order to garner more research. With a case study, enough detail is provided to truly observe what is happening because the researcher can identify mechanisms, causes and consequences (Flyvbjerg 2006, 220). The individual cases will consist of the best known U.S. presidential mistakes of the second half of the 20th Century. A mistake is a decision that resulted in extreme negative results or payoffs, either for the U.S. (i.e., loss of life or wealth with no positive results) or for the prestige of the president making the decision. U.S. presidents were selected because they are the single most powerful political figures in the U.S., if not the world. Mistakes are selected because they are closely covered by the press and historians so there is enough data to evaluate decision making using game theory and because the president appears to have selected an alternative that has the possibility of having the worst outcome. Such a decision is usually counter to the classical RAM. It would also be interesting to see if a decision, following classical RAM theory would not have resulted in a mistake, or at least the same mistake. In this study only ex-presidents from John F. Kennedy (JFK) on will be examined for the following reasons: 39 1) The more recent the president the more relevant the decision making process to current and future culture and context. Also, the president prior to JFK who made a large mistake, according to the McConnell Center study discussed later in this chapter, is considered to be Woodrow Wilson when he refused to compromise on the Versailles Treaty. Examining Wilson would pull us back in history over 85 years to World War I (Gregg, 2006). 2) The media were not as diligent in providing negative information on presidents prior to JFK. Even JFK's affairs, although well known by the press, went unexposed somewhat in the way that Roosevelt's wheel chair was not common knowledge at the time of his presidency. 3) Every president since (and including) JFK is best known for at least one disastrous decision or mistake. Table 3.6 lists some of these major mistakes. Table 3.6: Major Presidential Mistakes (1950-2000) President?s Name Mistake John Kennedy Bay of Pigs (botched invasion of Cuba) Lyndon Johnson Escalation of the Vietnam War Richard Nixon Watergate break-in and cover up Gerald Ford Pardon of Nixon (although this is arguable) Jimmy Carter Desert One (botched rescue of U.S. hostages in Iran Embassy) Ronald Reagan Iran-Contra Affair (sale of arms to Iran and providing profits to the Contras) George Bush Sr. Raising taxes after promising no new taxes Bill Clinton Sexual affair with intern Monica Lewinsky and cover up 40 In 2006, the McConnell Center, a nonpartisan leadership institute at the University of Louisville, conducted a survey to determine the worst presidential mistakes. Dr. Gary Gregg of the McConnell Center surveyed ninety political experts during the February 18-19, 2006 conference on ?Presidential Moments,? asking for what they considered the worst political mistakes of former presidents (Gregg, 2006). Appendix 1 provides a copy of the survey, a list of participants, and a summary of the results. The detailed results were never published and the summary was developed by this study?s author from the raw data provided by Dr. Gregg. Using the center?s list of top ten mistakes and this study?s criterion of only cases since (and including) JFK, five cases result. The cases are listed in Table 3.7. Table 3.7: Presidential Mistakes Making the McConnell Center?s Top Ten List and Occurring between 1961 and 2000 President?s Name Mistake Survey Ranking John Kennedy Bay of Pigs Invasion 8 Lyndon Johnson Escalation of the Vietnam War 3 Richard Nixon Watergate 5 Ronald Reagan Iran-Contra Affair 9 Bill Clinton Monica Lewinsky Scandal 10 (Source: Gregg, 2006) All of these mistakes will be used as cases in this study except the Iran-Contra Affair since it has never been proved that Reagan knew about the situation, let alone made a conscience decision about it (Walsh 1993, xiii). 41 Why did these supposedly intelligent men make decisions that either tainted otherwise relatively untarnished legacies or made them known primarily for their mistakes? Without Watergate, Nixon could have been remembered for ending the Vietnam War and for detente with the USSR and China. LBJ would be known for his Great Society and his stand against racism. Bill Clinton would be known for improving the economy. Yet all of these men, by their own actions, placed a cloud over their presidencies. Why did these presidential disasters happen and could they have been avoided? Game theory might give some answers, especially if informed by well known decision making models. Allison and Zelihow (1999) showed that all three of their models can often be applied to the same situation, but Allison and Zelihow were looking at both sides in a confrontation between countries (the Cuban missile crisis) which included many people and offices. They focused more on the events rather than looking specifically at JFK's decision. When looking at the decision of one man or woman, all three models may not be applicable. The primary model would depend on that person?s decision making style, as well as the available information surrounding the decision itself. Groupthink may also come into play here. The term groupthink was coined by Irving Janis (1972). He ascribes groupthink to dysfunctional groups that do not seek out or examine new alternative actions due to intense group cohesiveness. Once the group or group leader has hinted at what alternative he or she favors, the rest of the group members quickly come on board promoting and praising that alternative. No other alternatives are even suggested for fear of losing the support of the group. The classic example given by Janis is President Kennedy?s decision to go forward with the Bay of 42 Pigs invasion of Cuba, although he also applied it to Watergate and Johnson?s escalation of Vietnam (Janis 1972, 14-49). The game theory process is not very useful in a groupthink situation since the groupthink pathology makes it impossible to rationally consider a variety of alternatives. The groupthink concept will be identified in this study only if it pertains to a decision being analyzed. For each game (presidential decision), the alternatives are laid out and each alternative prioritized for each player according to their goals and objectives. The rules of the game are also laid out. All parts of the game are influenced by which decision making models are used. A RAM alone would follow the classical game theory approach. The OBM would restrict the goals and objectives to support of the organization while the GPM and EAM center on the objective of staying in power but differ on the mechanics used to accomplish this goal. The selection of alternatives is also different. Under the RAM approach, all alternatives are fair game, but OBM restricts alternatives to SOPs and GPM restricts alternatives to choices that would allow the winning of elections. This means that other players may choose alternatives not even considered by the player under study. In each case study, the situation will be described to include historical context and the actions of each of the major actors. Then, the event will be analyzed as a game using classical RAM theory where the alternatives and objectives of all players are listed. The RAM says that players want to maximize gains and minimize risks and that player A?s decision will be based on the assumption that player B?s decision will result in the worst outcome for player A. Under RAM, player A should pick the decision that gives him the best worst alternative (lesser of evils), no matter what player B decides. 43 In order to determine which model to use to inform the game theory analysis, four well respected books on presidential psychological analysis will be used. These books are by James Barber (1985), Alexander and Juliette George (1998), William Lammers and Michael Genouese (2000), and Fred Greenstein (2004). From these four books a consensus on each president's decision making style will be determined by summarizing what each book says is the ex-president?s style and comparing these summaries to the four different decision-making models. The model used will be the model that has the most traits in common with each book?s summary. After each decision is analyzed using game theory, it will again be analyzed using game theory informed by the appropriate decision-making model based on the ex- president's agreed upon decision-making style to determine if the alternative chosen can be better explained. If the four psychological books do not agree, and one book points to one model when a different book points to another model, or if the ex-president being looked at is an amalgamation of more than one model, then that case will be analyzed using more than one decision making model. The OBM says that players want to promote the organization, the player using the GPM wants to get elected, and the EAM player wants to stay in power. These decisions are made in accordance with the rules of the game as each player sees them. The matrix presented in Table 3.8 outlines each of Allison and Zelihow?s models, as well as the EAM. 44 Table 3.8: Key Elements of the Decision-making Models Applied to the Game Theory Process Rational Actor Model (RAM) Organizational Behavior Model (OBM) Governmental Politics Model (GPM) Elitist Actor Model (EAM) Goals/Objective ? Gain advantage over other players ? Win ? Grow the organization ? Stay in power ? Raise money ? Please constituents ? Win elections ? Retain power ? Grow power Alternatives ? Anything is possible ? Decision made by a single player in a deliberate manner ? Limited by o Organizational capabilities o Standard operating procedures o Organizational Culture ? Limited by alternatives that o Would win elections o Leave a legacy ? Limited to relying on other elites Outcomes/payoffs ? Straight forward ? Based on specific unit of measure or a prioritizing of alternatives ? Accomplish organizational mission ? Win elections ? Maintain position of leadership ? Move up in leadership positions ? Stay in power ? Retain power ? Grow power Strategy ? Minimize risk and maximize advantage ? Noncooperative with other organizations ? Small choices made by multiple players ? Cooperative ? Bargaining ? Noncooperative with other actors ? Secretive from other actors Rules of the game ? Depends on the game ? Keep the boss happy ? Take actions to get promoted ? Follow standard operating procedures ? Be a team player ? Loyalty to organization ? Legal rules ? Formal rules ? Informal rules, e.g., log rolling ? Cooperative only with other elites (Based on Allison and Zelihow (1999) and Dye (2001) and Mills (1956) It is important to select the most objective and detailed books available in determining the context and actions of each case. Books referenced and recommended by prominent historians specializing in the era will be used. Also, all major events must be consistent across every book used on the subject. Although books promoting a theory (such as who ?deep throat? really was from the Watergate era) can be used, any hypothesis or conspiracy theory promoted in the book will be ignored. 45 The next chapter presents the first case study, Kennedy and the Bay of Pigs. Subsequent chapters examine others in chronological order. 46 CHAPTER 4 JOHN F. KENNEDY: BAY OF PIGS All presidential mistakes have repercussions. Johnson expanded U.S. involvement in a long, demoralizing, and unnecessary war. Nixon and Clinton tarnished the credibility of the office of the president for generations to come. But before any of that happened John F. Kennedy made a mistake that was the primary reason that Khrushchev located a nuclear missile in Cuba, nearly precipitating World War III. John F. Kennedy, Democratic Senator from Massachusetts, became the 35th president in 1961 after winning an extremely close race against Richard Nixon. Kennedy was the first Roman Catholic and the youngest person (43) ever elected president. His presidency marked the beginning of the 1960s which was a decade where African- Americans, women, and homosexuals struggled to gain equal treatment and where the general population became disillusioned with government over the Vietnam War. Kennedy was assassinated in 1963, having served less than three years in office. Historical Context Fidel Castro came to power in January 1959 as the leader of a primarily middle class Cuban revolution against Fulgencio Batista who had allowed industry and education to deteriorate while using violence and corruption to stay in power (Lazo 1968, 58-72; Neustadt and May 1986, 141; Schlesinger 1965, 217). The Communist Party had assisted in Castro?s revolution, and as he awkwardly tried to form a government, the party gained 47 nearly complete control of the political process in Cuba (Lazo 1968, 206; Schlesinger 1965, 218). Although there were many democrats in Castro?s revolution, one by one, they were executed or jailed as traitors or they left the island until all democratic voices were silenced (Lazo 1968, 215; Schlesinger 1965, 219). As 1959 came to a close, the original goals of free elections, civil liberties, and land reform that brought Castro?s revolution to life and allowed it to succeed were trampled by Castro and the Communist Party (Lazo 1968, 229; Schlesinger 1965, 220). Before Castro came to power U.S. politicians were unsure whether to support him or Batista, but once Castro came to power he was quickly recognized as the legitimate leader of Cuba (Schlesinger 1965, 220). When Castro visited the U.S. in April of 1959, he was met with a warm welcome, offers of economic assistance, and requests to compensate American investors for the nationalization of their property, but Castro rejected all offers and requests (Lazo 1968, 228-229; Neustadt and May 1986, 141). After returning from his visit, he represented the U.S. government as hostile to Cuba?s autonomy and national security and used this position as an excuse to forgo national elections (Lazo 1968, 215; Schlesinger 1965, 221). U.S. policy makers made several attempts to reach an arrangement with Castro but were continually rebuffed. Finally, in 1960 the U.S. refused to buy Cuban sugar and then broke off all diplomatic relations in January 1961 when Castro forced the staff of the U.S. Embassy in Cuba to depart the country (Lazo 1968, 227; Schlesinger 1965, 222). The idea for what would eventually become the Bay of Pigs operation was developed by the Central Intelligence Agency (CIA) in December of 1959. The CIA feared Fidel Castro?s growing popularity. He was a strong communist who led the 48 country to overthrow a right wing dictator who had been enjoying U.S. support. At a December meeting, the CIA decided to develop a plan to organize opposition to Castro (Wyden 1979, 19). In January 1960 the plan was refined to mirror an operation in Guatemala where the CIA, with only 150 Guatemalan exiles, had overthrown the old communist government and established a new government using a radio station for propaganda, and a few Air Force bombers (Gleijeses 1995, 7; Wyden 1979, 20). The Guatemala operation was so highly successful, lasting only a week and costing only one life, that the CIA was confident that the same thing could be done in Cuba (Wyden 1979, 21). The few people in the CIA who disagreed with the Cuban operation would only distance themselves from it rather than protest since they felt they would not be taken seriously due to this atmosphere of overconfidence (Wyden 1979, 34). A U.S. naval base outside Miami, Florida was activated for the operation, and many of the old crew that ran the Guatemala operation were assigned to the new Cuba one (Schlesinger 1965, 227). The first task was to set up a radio station. Although the studios were in Florida, the broadcast could not be traced back to the U.S. because the transmitter was installed on Swan Island, an almost deserted location off the coast of Honduras, which had been under disputed ownership by the U.S. and Honduras since 1863 (Higgins 1987, 24; Wyden 1979, 23). Before spring of 1960 Radio Free Cuba, known to the CIA as Radio Swan, was up and operating (Gleijeses 1995, 6). At the same time that planning for the Bay of Pigs operation was underway the CIA was working on another operation that could make the Bay of Pigs unnecessary. This was the plan to assassinate Castro (Higgins 1987, 55; Wyden 1979, 23). 49 President Eisenhower never saw the plan to assassinate Castro, but on March 17, 1960 he approved a 13 million dollar CIA program for Cuba (Gleijeses 1995, 10-12; Lazo 1968, 264; Neustadt and May 1986, 142; Schlesinger 1965, 222-228; Wyden 1979, 25) which consisted of the following four efforts: 1. Gather a group of acceptable Cuban exiles that could step in to quickly form a new Cuban government. 2. Use propaganda to try to set Cubans against Castro. 3. Organize a democratic underground with intelligence gathering capability inside of Cuba. 4. Organize and train a paramilitary force of Cuban exiles outside Cuba. By May of 1960 the CIA was training 20 radio operators on a farm in Guatemala (Schlesinger 1965, 236). The purpose of these operators was to transmit coded intelligence messages from within Cuba (Gleijeses 1995, 6; Wyden 1979, 36). They were to be the communication link between the underground organization in Cuba and the CIA. Eventually others were added who would gather intelligence and organize a resistance movement (Schlesinger 1965, 228; Wyden 1979, 38). The paramilitary force was also being trained in Guatemala and by the end of 1960 numbered about 500. All participants recognized that this would be woefully inadequate against Castro?s army of 200,000 men (Wyden 1979, 56). The military and CIA were counting on aerial bombing support and a national uprising to equalize the odds (Wyden 1979, 64). By December 1960 the plan was refined further and called for 3,000 Cuban exiles, although only 700 were being trained, to land on the beach after air strikes cleared 50 the way for them (Wyden 1979, 69). This expansion of the plan from a guerrilla operation to an invasion was not approved by anyone outside of the CIA. Cuban pilots were trained to fly B-26s because they had been sold as excess inventory to a variety of countries and were not easily traceable to the U.S. The hope was that the operation would have plausible deniability by the U.S. government (Lazo 1968, 271; Wyden 1979, 70). The U.S. military was very reluctant to help the CIA which needed them to provide planes and trainers. Military officers believed that the plan was unworkable because the invasion force was too small and there was no popular leader who would motivate the rest of Cuba to join the fight (Higgins 1987, 65; Wyden 1979, 71). Castro had already executed most of the leadership of his opposition. The few men around whom Cubans could rally were excluded from the CIA operation as too independent ? the CIA needed to control the operation and free thinkers were an inconvenience (Lazo 1968, 215; Schlesinger 1965, 231; Wyden 1979, 72). In February 1961, 36 Cuban exiles were secretly sent to Cuba to gather intelligence and encourage an underground force opposed to Castro which the U.S. would arm (Wyden 1979, 75). Around that same time the CIA contracted with a small Cuban shipping company called the Garcia Line Corporation, which owned six freighters, to transport the Cuban exiles to the Bay of Pigs (Wyden 1979, 77). The plan was for the B- 26s to destroy the Cuban Air Force and then two of them would fly to Miami to defect to give the illusion that Cuba was in civil war (Higgins 1987, 131; Schlesinger 1965, 229; Wyden 1979, 163). Once the Air Force was destroyed the invasion could begin. The objective was to capture and defend a piece of real estate that was large enough to accept 51 defectors and freedom fighters. It would become a staging area where the uprising could begin and soon spread throughout the island. A provisional government could then be flown in, which the U.S. would recognize and support (Gleijeses 1995, 18; Lazo 1968, 275-277; Schlesinger 1965, 234; Wyden 1979, 171). John F. Kennedy and the Bay of Pigs Kennedy while running against Nixon for president in 1960 expressed the view that the U.S. had made policy decisions in Cuba which promoted the rise of Castro. Kennedy believed that the Eisenhower administration had turned its back on the people of Cuba by supporting the corrupt and repressive Batista in order to promote the interests of American companies in Havana (Higgins 1987, 58; Schlesinger 1965, 225). Kennedy thought that the U.S. had broken faith with the Cuban people and that was why the Communists could so easily gain a foot hold in Cuba. His only solutions at the time, were to support any democratic uprising in Cuba and to provide other countries the support we should have shown to Cuba so Communism would not gain strength in other countries (Schlesinger 1965, 225). During the Nixon-Kennedy presidential debates, Kennedy took the position that the U.S. should help Cuban rebels and exiles rise up against Castro. Nixon was forced to take the opposite position even though he was a strong proponent of the CIA operation in order to not jeopardize the secrecy of the operation (Neustadt 1990, 244; Neustadt and May 1986, 142; Schlesinger 1965, 225). Nixon was persuasive in arguing his side, saying that such an action would be illegal and condemned by the international community and would encourage the USSR to interfere in Latin America (Wyden 1979, 67). 52 Nixon was judged to have lost the debates, perhaps in part because his Cuban position did not resonate with the American public (Wyden 1979, 68). At the same time, Kennedy?s position may have had the effect of making it more difficult for Kennedy to stop the CIA operation later (Wyden 1979, 68). Soon after Kennedy won the election he was briefed on the CIA?s Cuban operations (Husain 2005, 24; Neustadt 1990, 244; Neustadt and May 1986, 142). Kennedy did not raise any objections at that time, but in January 1961 he held a meeting with his cabinet on the subject at which Kennedy asked the Department of Defense (DOD) to evaluate the CIA plan and the State Department to develop a plan to isolate Cuba with the help of the Organization of American States. The CIA was allowed to continue its planning as long as it continued to appear to be a Cuban exile operation only (Gleijeses 1995, 22; Schlesinger 1965, 238). To maintain secrecy, Kennedy limited the number of DOD personnel allowed to examine the plan. Due in part to this restriction, the DOD evaluation was ambiguous at best. At one point it declared the CIA plan unworkable without significant additional support from either the Cuban people and military or the U.S. military yet it said that the overall plan had a ?fair? chance of success without reiterating the previously stated conditions. The word ?fair? was misinterpreted as giving the plan a reasonable chance at success when that is not what the author intended (Gleijeses 1995, 26; Husain 2005, 25; Neustadt 1990, 244; Neustadt and May 1986, 142; Schlesinger 1965, 238). By March 1961 several conditions developed making it necessary for Kennedy to make a decision on the CIA operation. The training base in Guatemala was becoming more and more difficult to hide as locals continued to observe activities there and the 53 government of Guatemala asked that the CIA close up shop (Neustadt 1990, 244; Neustadt and May 1986, 143; Schlesinger 1965, 239). Also, the spring rainy season would make training impossible. Finally, intelligence sources had confirmed that in June the USSR would provide Cuba with jet fighter aircraft that 17 B-26s would be unlikely to destroy prior to the landing. Kennedy had to let the CIA execute its plan by June or give up the chance altogether (Higgins 1987, 91; Schlesinger 1965, 240). Because the CIA had proposed the Cuban operation and had worked on it for over a year, CIA leaders were heavily invested in seeing it implemented. They were staking their reputations on its success and strongly advocated its execution to the newly elected president (Neustadt 1990, 244; Schlesinger 1965, 240). In fact, they pushed it so strongly as to point out that the Cuban exiles currently being trained would present a security risk if they were not allowed to execute the plan since they would be demoralized and would likely reveal the classified operation to their kinsmen (Husain 2005, 24; Neustadt and May 1986, 143; Schlesinger 1965, 242). This, the CIA emphasized, would make the U.S. look impotent and would encourage more communist excursions into Latin America. The CIA made it appear that it was imperative that their operation go forward (Schlesinger 1965, 252). Kennedy wanted to know why the Cuban exiles that were to storm the beaches and fly the B-26s could not be used as spy and underground organizers instead. CIA briefers agreed that they could. But no one ever considered just debriefing the exiles by warning them that they could go to jail if they disclosed classified operations (Schlesinger 1965, 242). Based on these CIA concerns, Kennedy elected to let the CIA execute its plan as long as there was no U.S. military involvement. None of his cabinet and staff raised any 54 objections or concerns. Kennedy summed up the meeting by saying that there must be a balance between political risk (letting the world find out that the U.S. was behind the operation) and military risk (not winning). Kennedy also reserved the right to call off the operation within 24 hours of its execution (Neustadt and May 1986, 144; Schlesinger 1965, 243). Kennedy gave himself an out because he was unsure how unpopular Castro was in Cuba and whether the Cuban people would really rise up against him (Schlesinger 1965, 246). The CIA seriously failed Kennedy. It never produced an intelligence estimate of how likely an uprising would be although when pressed, the planners speculated that over one quarter of the Cuban people would provide active support (Schlesinger 1965, 247). If the intelligence arm of the CIA or the State Department had been invited to evaluate the CIA planners? statement (which they were not) (Neustadt and May 1986, 143; Schlesinger 1965, 248), they would have pointed out that many young people supported Castro since they found themselves better off under his regime and those who did not support Castro would still not support a U.S. invasion since the U.S. had supported the prior Batista regime that many believed was worse (Higgins 1987, 91; Neustadt 1990, 246; Schlesinger 1965, 249). The briefings to the president seemed to indicate that an invasion would quickly spark a general uprising even though the CIA was not unified in its belief that this would happen. The CIA?s plan actually depended on the exiles establishing a base camp from which the current Cuban resistance could operate. Here lies a subtle but critical difference in expectations between an immediate uprising that Kennedy was expecting and the more drawn out affair that the CIA was really planning. The president would 55 probably have not supported a plan where 1,000 men had to establish a beach head against 200,000 soldiers for an extended period of time (Gleijeses 1995, 35; Higgins 1987, 103; Lazo 1968, 277; Schlesinger 1965, 247). As Kennedy showed more and more concern about the operation the CIA assured him that if the deployment failed the exiles could just escape into the mountains and join the resistance. But the CIA did not explain, and no one asked, if the exiles had guerrilla training (most did not) and how far away the mountains were (80 miles, across jungle and swamp). Also, the Cuban exiles were never told to escape to the mountains. They were only told (until right before the execution of the operation) that if they failed the U.S. military would rescue them (Gleijeses 1995, 39; Neustadt 1990, 246-247; Neustadt and May 1986, 143; Schlesinger 1965, 250). Only two people ever raised an objection directly to Kennedy. One was Arkansas Senator William Fullbright, Chairman of the Senate Foreign Relations Committee. Fullbright made a convincing argument against invasion at the end of March. He pointed out that an invasion would go against the Organization of American States Charter and other treaties. If it succeeded, we would be committed to ensuring the success of the U.S. formed Cuban government, and if we committed U.S. troops to its success, we would be seen around the world as deposing a legitimate government and hypocritical, given prior objections to the USSR doing the same thing in other locales. The U.S. would lose the moral high road. Fullbright pressed for Kennedy?s initial solution of isolation of Cuba (Lazo 1968, 279; Neustadt and May 1986, 145; Schlesinger 1965, 251; Wofford 1980, 345). 56 Arthur Schlesinger, Special Assistant to the President, also raised a more practical objection very late in the decision process. He believed that the U.S. would not be able to hide its involvement, and once the invasion began it would have to commit American troops in order to hold a beach head and that once a beach head was established would have to commit even more troops or endure heavy losses along with a prolonged conflict (Neustadt 1990, 244; Schlesinger 1965, 253). If the U.S. committed enough troops to quickly win, it would lose face with the rest of the world by making an unprovoked attack on a sovereign nation (Schlesinger 1965, 254). Under Secretary of State Chester Bowles also had objected but not directly to Kennedy. He had once sat in on one meeting for Secretary of State Dean Rusk and was horrified by the plan but did not think it was his place to speak for Rusk. When next he saw Rusk he urged him to convince the president that the operation was a grave mistake. Rusk did not pass on Bowles? concerns, and Bowles was excluded from future meetings (Higgins 1987, 107; Wofford 1980, 344). Making the operation appear to be a civil war with no U.S. involvement seriously endangered the operation. The U.S. military pointed out that there were not enough exiles, pilots or B-26s to make the operation militarily feasible against Castro?s army of 200,000. U.S. military forces would make it a viable operation, but that could open the door for the USSR to send troops in support of Castro, therefore escalating the conflict into a war between the U.S. and USSR if it did not end quickly. But of almost as much importance, this was an illegal operation with one country invading another. It went against everything the U.S. publicly stood for and would not be viewed favorably by world opinion (Schlesinger 1965, 239; Wyden 1979, 164). 57 Based on CIA assurances, Kennedy viewed the operation as more an infiltration effort than an invasion, yet he did not ask the CIA to redirect the plan to infiltration only (Higgins 1987, 97; Schlesinger 1965, 257). The CIA assured Kennedy that the 800 Cuban exiles wanted to free Cuba one way or the other, and Kennedy thought it was safe to let them try after all the hard work and training they had undergone. Kennedy decided that if the Cuban exiles wanted to try to free Cuba and they understood that they would get no U.S. military help, then they were welcome to try (Higgins 1987, 115; Schlesinger 1965, 258). Kennedy did not see a downside. If they succeeded, he would be free of a Communist government 90 miles off the coast of the U.S. If they failed, it would not be traced back to the U.S. since all the Cuban participants would be dead or would have melted into the countryside to join the resistance (Higgins 1987, 122; Husain 2005, 25; Schlesinger 1965, 258). Kennedy had decided not to use U.S. military forces no matter what happened because he was adamant that the operation not be traced back to the U.S. The success of the entire operation depended on the people of Cuba rising up against Castro at the critical moment when the exiles landed on the beach of the Bay of Pigs?an event that could not be predicted and seemed almost like wishful thinking. There might have been a chance if the CIA had worked with the Cuban exiles as partners in achieving a goal, but instead they treated them as puppets leaving the Cuban people without their own organization or leadership for revolt. In the mean time Castro had interned 100,000 of his population that he did not trust in order to discourage any thoughts of a counterrevolution. The exiles for their part never believed the U.S. would abandon them on the beach without U.S. military back up?even when they were told of Kennedy?s 58 decision. They always thought the U.S. would support them rather then allow them to fail (Neustadt and May 1986, 144; Schlesinger 1965, 231; Wyden 1979, 164). Kennedy?s decision may have made sense if the information the CIA provided had been accurate but the CIA had deceived him and neither he nor his cabinet ever asked the tough probing questions that would have forced the CIA to admit the truth (Gleijeses 1995, 13; Neustadt and May 1986, 144; Schlesinger 1965, 258). Part of the reason for hesitancy in asking the tough questions may have been that most of the participants were new to their jobs, and part of the reason may have been that it was Richard Bissell, head of the CIA?s Directorate for Plans, who was pushing the plan. Bissell had a strong personality and an irreproachable reputation as he had been a driving force behind the success of the U-2 spy plane and the spy satellite programs (Lazo 1968, 279; Neustadt and May 1986, 144). On April 17, 1961, 1,400 Cuban exiles attempted an amphibious assault on Cuba via the Bay of Pigs. Six B-26s and a few camouflaged B-29s were able to disable over half Castro?s Air Force. The undamaged Cuban planes destroyed two of the four ships carrying ammunition, communications gear, and other supplies. The other two ships ran for open waters. The Cuban Air Force kept the Cuban exiles on the beach pinned down until Castro?s army arrived. Although seven U.S. Naval destroyers lay 30 miles off Cuba, Kennedy would not allow them to take any action (Wofford 1980, 350). By the third day, all the exiles were either dead or in jail (Higgins 1987, 129; Janis 1972, 15; Neustadt 1990, 246). The landing on the Bay of Pigs was a dismal failure. There was no uprising and no beach head established. Not only were the Cuban exiles quickly killed or captured 59 with none making it to the mountains, but the operation was just as quickly traced back to the U.S. Kennedy had to endure criticism at home and abroad impacting his ability to reach a compromise with the USSR and leading to another crisis in Cuba the following year over the set up of a Soviet nuclear missile site. (Husain 2005, 26; Janis 1972, 15; Lazo 1968, 302; Neustadt and May 1986, 144; Schlesinger 1965, 267). In 1962 an organization called Tractors for Freedom, unofficially backed by the Kennedy administration and led by Eleanor Roosevelt, gave Castro $53,000,000 worth of food and medicine in exchange for the release of the captured exiles (Schlesinger 1965, 839). Janis looks at the Bay of Pigs fiasco as a perfect example of what he terms groupthink behavior. According to Janis, Kennedy and his staff made six incorrect assumptions which led to Kennedy?s colossal mistake (Janis 1972, 19). These assumptions are: 1. The Bay of Pigs attack will not be traced back to the U.S. 2. The Cuban Air Force can be completely destroyed by a few old B-26s from World War II. 3. The exiles will continue to fight once they realize the U.S. is really not going to provide back up. 4. Fourteen hundred exiles can retain a beach head against 200,000 soldiers. 5. An invasion of 1,400 exiles will set off an uprising. 6. If unsuccessful, the exiles can melt into the mountains and join the underground. (Janis 1972, 19-29). Janis first examines what might be termed official or conventional explanations for why no one asked the probing questions needed to debunk these six bad assumptions. 60 This official explanation drawn from various individuals present at the time boil down to four intervening factors. The first factor was ?political calculation.? This factor centered on the issue of what to do with the exiles if they were not sent to Cuba and the perception that this was Kennedy?s one and only chance at getting rid of Castro before new Soviet planes arrived (Janis 1972, 32). The second factor was that this was a brand new administration. Kennedy was an inexperienced president and his advisors did not feel ready to deal frankly and bluntly with Kennedy or each other. The third factor was the limiting of expert input due to the perceived need for secrecy. The final factor was the fear of being the lone objector; one person does not want to stand out as being negative or ill informed if all the other advisors appear to be satisfied with the decision (Janis 1972, 31-33). Wofford agrees with these factors but also adds that Kennedy?s impromptu and unstructured decision making style may have been a major factor. Kennedy would not discuss one issue in depth but would jump around. He would become impatient with long responses and would sometimes ignore chain of command (Wofford 1980, 355). Janis discounts these factors as not providing a full explanation of such a poor decision. Although Kennedy was a new president, he was not new to politics and many in his cabinet and staff had years of experience with policy matters and with dealing with each other (Janis 1972, 33-35). Janis sees groupthink, a term that he invented, as the only explanation of why some of the best minds in the country could have made such a blatant error. In groupthink a small group maintains its cohesiveness and high morale by believing that the group is invulnerable, especially if its numbers reach a unanimous conclusion. Once such a conclusion is made no one wants to express any doubts for fear 61 of being alienated from the group or losing group solidarity. The group feels almost magically omnipotent as long as this state is maintained. Such a situation makes it difficult to make logical, rational decisions, resulting in the six faulty assumptions that gave rise to the Bay of Pigs mistake (Janis 1972, 38-49). Dixon provides a similar explanation for the mistake. He observes that there is often an ?unrealistic over-confidence in rapid victory? not only by those who gravitate toward the military but by governments that attempt to wage war by using committees to make military decisions (Dixon 1976, 397). Rational Actor Model At first blush one would think that the Bay of Pigs game was essentially a two player game between Kennedy and Castro. At the time of the Bay of Pigs action the only prior moves were between the president (first Eisenhower and then Kennedy) and Castro and were of an economic and diplomatic nature. This game was a consecutive move game where Castro would nationalize U.S. companies or close the U.S. Embassy in Havana and the U.S. would respond or the U.S. would stop buying Cuban sugar and Castro would respond. There were many alternatives discussed at the time for dealing with Castro such as assassination, invasion, economic sanctions, diplomatic sanctions, supporting a Cuban underground, or doing nothing. The Bay of Pigs game was more specific. It had only to do with what decision Kennedy would make on a CIA operation which eventually came to be known as the Bay of Pigs. This game was not a game played against Castro because there was no question of how Castro would respond. His antagonism against the U.S. was one reason for trying to overthrow him. 62 In Kennedy?s mind, the player who would make or break a Cuban exile invasion was the Cuban people themselves. The solution to the Bay of Pigs invasion game was determined by how the Cuban population would react. It was a game played against the Cuban people because their decision of whether to rise against Castro or not was the key to the success or failure of the Bay of Pigs operation. In this game, the only alternatives Kennedy could make were the following: Kennedy Alternatives: Cancel operation Redirect operation to infiltration only Execute operation This was a consecutive game based on the Cuban population reacting to Kennedy?s move. Once Kennedy took action, then the Cuban people were viewed as having the following alternatives: Cuban Alternatives: No uprising Slow uprising Quick uprising A game matrix developed from these alternatives is shown in Table 4.1. Table 4.1: Bay of Pigs Game Theory Matrix Cuban People No Uprising Slow Uprising Quick Uprising Cancel Operation Redirect to Infiltration Kennedy Execute Operation 63 The rules of the game as Kennedy saw it were that the Bay of Pigs game was a one time consecutive game, communication between players flowed one way, from the U.S. to the Cuban people via Radio Free Cuba and agreements between players were impossible. Both players could make only one move with Kennedy going first. The problem with this game is that Kennedy?s perception of the priorities of the Cuban population was vastly different from their actual priorities. Kennedy interpreted the game as negative sum if the deployment was cancelled while it would actually be zero sum since most of the Cuban people were not interested in a counterrevolution and those that were were incarcerated. From accounts of his discussions with his staff and others, Table 4.2 below describes how Kennedy probably viewed the payoffs of different alternatives. Benefits are listed above the dotted line under each column, and costs are listed below the dotted line. 64 Table 4.2: Payoff of Different Options for Kennedy Concerning the Bay of Pigs Operation1 Cancel Redirect to an Infiltration Mission Execute No confrontation with USSR No casualties ......................... Communist aggression may escalate in Latin America U.S. will look weak and impotent Kennedy would not be standing by his campaign rhetoric and would look weak and impotent Exiles could leak the classified mission and operations will be traced back to U.S. Exiles are utilized and will not leak classified mission Could contribute to eventual overthrow of Castro Low risk of casualties Low risk of confrontation with USSR Low risk of operation traced back to U.S. ........................ Castro will not be quickly removed Castro could be quickly overthrown and a democratic Cuba established Exiles are utilized and will not leak classified mission Anti-communist proponents in U.S. will be pleased ......................... High risk of casualties High risk of confrontation with USSR High risk of operation traced back to U.S. 1It could be argued that there is a fourth alternative, that of execution with direct U.S. military support. Such an alternative could quickly progress into an outright U.S. invasion of Cuba which would be a totally different game since an uprising of the Cuban people would no longer be as crucial, making Castro, not the Cuban public, the second player. However, participant accounts suggest this alternative never received serious consideration. Table 4.3 below describes how the Cuban people probably viewed the payoffs of different alternatives. Benefits are listed above the dotted line under each column, and costs are listed below the dotted line. 65 Table 4.3: Payoff of Different Options for the Cuban People Concerning the Bay of Pigs Operation No Rise Up against Castro Slow Rise Up against Castro Quick Rise Up against Castro Low risk of being put in jail or executed Possibility of slow improvement in quality of life ......................... No elections Little freedom Possibility of winning a democratic government More confident in numbers available to win Less risk of jail or execution ......................... Long drawn out civil war Possibility of winning a democratic government ......................... High risk of being put in jail or executed For Kennedy, the alternative with the worst payoffs would be execution of the Bay of Pigs operation, whether the exiles lost or won. Winning in Cuba could result in a larger war with the USSR, which could be catastrophic if it developed into a major war where nuclear weapons were used. This almost happened a year and a half later even though the exiles lost. Losing in Cuba could result in loss of lives, money, and countries to communism. The two alternatives with the best payoffs would be cancellation and redirection. Assuming that the Cuban people select the alternative that gives Kennedy the worst payoff, then they would not rise up. A rational actor would always hope for the least amount of effort for the best result. Kennedy would prefer the other player to select payoffs in the following order: quick rise, slow rise, no rise. The Cuban people on the other hand would hope for Kennedy to cancel, redirect and lastly execute. Based on these preferences, a prioritization of outcomes or results can be developed. For example, if Kennedy was 66 making a decision purely as a rational actor, his best outcome would be for the U.S. to not do anything while Cubans quickly rise up and overthrow Castro (cancel, quick rise). This would be the least amount of effort for the best result and would therefore be Kennedy?s number one priority. For the Cuban people, their lowest priority would be for Kennedy to execute the plan and for them to have a quick uprising (execute, quick rise). Interestingly enough, the one thing that Kennedy actually did and was hoping would happen (execute, quick rise) was the absolute lowest priority for the Cuban people under a Rational Actor Model (RAM). Basing the matrix values on this RAM prioritization, all players, if they were rational actors, would probably prioritize like Table 4.4 below. Table 4.4: RAM Prioritization for Kennedy-Cuban People Kennedy Priority Result Cuban People Priority 1 Cancel ? Quick Rise 7 2 Redirect ? Quick Rise 8 3 Execute ? Quick Rise 9 4 Cancel ? Slow Rise 4 5 Redirect ? Slow Rise 5 6 Execute ? Slow Rise 6 7 Cancel ? No Rise 1 8 Redirect ? No Rise 2 9 Execute ? No Rise 3 Based on the above prioritization the RAM matrix payoffs could therefore look like Table 4.5 below. 67 Table 4.5: Bay of Pigs Game Theory Matrix Cuban People No Uprising Slow Uprising Quick Uprising Cancel Operation 7 , 1 4 , 4 1 , 7 Redirect to Infiltration 8 , 2 5 , 5 2 , 8 Kennedy Execute Operation 9 , 3 6 , 6 3, 9 Under the RAM, Kennedy should assume that the Cuban people would choose no rise to give Kennedy his worst case scenario. Then, Kennedy should choose the cancel alternative to give him his best worst case scenario. One problem is that Kennedy perceived that a quick rise had no chance of occurring without a catalyst such as the Bay of Pigs invasion. So Kennedy thought that without the execute alternative, the quick rise alternative was extremely unlikely. Another problem is that Kennedy had blended the execute and redirect alternatives into one to the point where selecting execute also meant selecting redirect even though, in reality, this was not the case. In Kennedy?s mind, he had only two alternatives ? go or no go. If the execute alternative did not cause a quick rise, the redirect alternative might cause a slow rise. But in spite of all this, the no rise scenario was his worst case scenario and cancel would give him the best outcome under a no rise situation. Determining a Different Utility Function Since Kennedy?s Bay of Pigs decision did not follow the classic RAM, he must have either made his decision using a different utility function or he made a purely emotional decision that bore little resemblance to well considered rational thought. The 68 first step in determining what his decision was based on is to look at what some authors of in-depth psychological studies believe were Kennedy?s goals, objectives, and driving forces. Lammers and Genovese (2000) paint a picture of a man who was a heroic figure and who used ?lofty rhetoric? to motivate the population. He was very charismatic and promoted an optimistic view of politics. He worked effectively with other politicians preferring to use persuasion over bargaining or bullying. In spite of the fact that deliberations over the Bay of Pigs operation seemed to discourage dissent, Kennedy seemed to want a wide range of input since his cabinet and advisors were very diverse in age and background (Lammers and Genovese 2000, 187-213). From this analysis, it would seem that Kennedy does not follow the rational actor, governmental politics, organizational behavior, or elitist actor models to any significant degree. Greenstein (2004) sees Kennedy as an inspirational figure who surrounded himself with brilliant advisors who were devoted to him (Greenstein 2004, 59-74). In spite of his talented advisors, Kennedy?s informal and unstructured method of deliberation made rational decision making difficult. Kennedy was a passionate and emotional man who appeared to live every day to its fullest whether he was trying to stop Castro, negotiate with the Soviets, or seduce a Hollywood starlet (Greenstein 2004, 59- 74). Again, Kennedy emerges as the exception to any model ? too dedicated to his principals to be governed by RAM or governmental politics model (GPM), too unstructured for organizational behavior model (OBM) and to respectful of others to be governed by EAM. 69 George and George (1998) describe Kennedy?s management style as ?collegial.? They show him as a confident leader who actively participated in the political process. He built strong working relationships through teamwork and a shared sense of purpose. Although his process was usually informal and unstructured, it tended to foster honest debate (although not in the Bay of Pigs case) and avoid intimidation by higher ranking advisors (George and George 1998, 210). Barber (1992) describes Kennedy as a man who inspired action and idealism in others. He gave his staff and advisors the feeling that they could accomplish anything and nothing was off limits. Barber posits that Kennedy had an active-positive personality which means that he was personally committed to everything he tried to accomplish and that he tried to accomplish a good deal. Failures were to be learned from, not dwelled on (Barber 1992, 361). Barber (1992) characterizes all presidents into four personality types in order to compare and contrast presidential personalities. The four types are based on two characteristic spectrums: ?activity-passivity? and ?positive-negative affect.? Activity- passivity is based on level of energy. A president who is always working is considered active while one that who takes much time off is considered passive. Positive-negative affect has to do with how much a president enjoys his life and feels satisfaction in it (Barber 1992, 8). Barber believes that these two personality dimensions can describe four basic and unique character types because they represent key aspects of everyone?s orientation on the world. The four types are active-positive, active-negative, passive- positive, and passive-negative (Barber 1992, 9). 70 According to Barber, as an active-positive Kennedy exhibited high energy, drive, humor, and passion for everything he did. He was competitive when running for office because he loved winning (Barber 1992, 384). He also did something few politicians ever did. He expected people to support him because they believed in his policies, not because they wanted positions in his administration or government contracts for their company (Barber 1992, 357). George and George and Barber still do not help place Kennedy in one of the standard models. Clearly Kennedy wanted to win, wanted to stay in power, but in the final analysis he stuck to his beliefs no matter what. A Different Utility Function In the final analysis Kennedy did not match any model. At first glance his decision seemed to be based on the OBM since Kennedy?s decision appeared to be heavily influenced by the CIA, but Kennedy?s mistake was not made because he supported the organization of the CIA, it was made because he did not question their information. His utility function was winning and doing what he thought was the right thing. Using this utility function drastically changed Kennedy?s priorities in the Bay of Pigs game. He still wanted the other player to have a quick up rising, then slow uprising, and then no uprising. The problem is that the Kennedy utility function of winning and following his beliefs made it very difficult for him to cancel the operation. Castro was continually goading the U.S. and had strong alliances with the USSR and other communist nations. Most Americans wanted something done about him and Kennedy thought that the Cuban people, who initially supported Castro only because they thought they were getting a democracy, no longer wanted him in charge. Also, it is not unusual for Kennedy to view ethical dilemmas from a dichotomous point of view. It is normal for 71 anyone to look at this kind of situation as having only two alternatives: to either take the proposed action or not take it. When faced with this type of dilemma other solutions are often not explored (Cooper 1998, 23). If Kennedy could not cancel because of his desire to win and support the Cuban people, and because of his inability to explore other solutions, then the matrix would look as shown by Table 4.6 below. Table 4.6: Bay of Pigs Game Theory Matrix under Kennedy Utility Function Cuban People No Uprising Slow Uprising Quick Uprising Redirect to Infiltration 5 , 1 3 , 3 1 , 5 Kennedy Execute Operation 6 , 2 4 , 4 2 , 6 If the cancel alternative were deleted, then the worst alternative is redirect, which probably would have more than satisfied Kennedy?s utility function of winning and doing the right thing plus the entire incident would not have blown up in his face if the exiles were smuggled in without being noticed. In fact Kennedy talked at length to the CIA about the exiles sneaking off to the mountains if they could not establish a beach head. Because of this Kennedy saw the execute and redirect alternative as one while the CIA knew that the exiles would never make it to the mountains. Nobody, not even Kennedy, could make a rational decision without accurate information, and whether due to groupthink, inexperience, or some other reason, Kennedy did not aggressively question what the CIA was telling him about the Bay of Pigs operation. Yet, to some extent, at least by his own perception of his choice, he did make a decision that fit in with RAM using his own utility function. 72 Conclusion In this case study, use of the RAM does not explain Kennedy?s decision to execute the CIA?s Bay of Pigs operation, but a modified utility function helps to explain the decision based on the fact that Kennedy was poorly informed by the CIA about his alternatives and the motivations of the other player. RAM can not work without at least a reasonable amount of accurate information. Taking into account what Kennedy believed about the operation he still did not make a rational decision about the operation unless you also take into account that execute and redirect were the same decision to Kennedy. Assuming a worst case decision of no uprising, cancellation would be the RAM choice and redirecting to infiltration would be the predicted RAM choice for Kennedy?s personality. Had the CIA provided accurate information or had Kennedy asked the right questions of the right people, it is very likely that the Bay of Pigs mistake would have never happened. 73 CHAPTER 5 LYNDON JOHNSON: ESCALATION OF VIETNAM Lyndon Johnson was John F. Kenney?s Vice President. He came into office when Kennedy was assassinated on November 22, 1963. The following year Johnson ran for president and was elected by over 60 percent of the popular vote. His primary goal was to eliminate poverty in the United States. Instead he embroiled the country in a war that most agree was devastating for the United States and should have been avoided. Historical Context For over a thousand years Vietnam was under the control of China. The people of Vietnam won their freedom in the 10th century after a long bloody war against Mongol forces. Then in the mid 1800s, as part of the European expansion and colonization, Vietnam became a colony of France. During World War II, Japan occupied Vietnam, using it to launch military attacks on nearby countries. At this time, an underground group with Communist leanings organized to combat the Japanese. This underground organization grew and refocused against the French when they came back to reclaim their colony after World War II. The resulting war of independence between France and Vietnam lasted until the French evacuated the country in 1954. This marked the first time the West suffered a defeat at the hands of a Communist movement. With the departure of the French, an international meeting, called the Geneva Accords, was held to determine the fate of Vietnam. Since there was conflict over who represented the 74 leadership in Vietnam, the Geneva Accords divided the country at the 17th parallel until elections could be held in 1956 and the country reunified (Holmes 2001, 953). Elections were never held. The United States never endorsed the Geneva Accords. Instead, President Dwight Eisenhower chose to support the government of South Vietnam which was a remnant of the government established by the French because he was convinced that a free election would result in a Communist government for Vietnam. The U.S. also supported the establishment of the Southeast Asia Treaty Organization (SEATO) which was an alliance of independent non-Communist countries (Hess 2001, 77). This U.S. intervention prevented free elections, even though this was contrary to agreements reached in Geneva, because the U.S. feared that the communists would win. The popular belief at the time was that every country lost to communism would encourage more countries to convert causing the U.S. and its allies to lose influence and access to natural resources. This idea, referred to as the domino theory, was not fully embraced by the intelligence community (Holmes 2001, 953). For the North, reunification was the overriding objective. To that end the North supported a growing insurgency in the South in order to destabilize the government there. The North provided supplies and manpower to the insurgents, known as the Viet Cong, via primitive roads through mountains and jungles between the North and the South. These routes became known as the Ho Chi Minh Trail (Hess 2001, 78). President John Kennedy became involved with Vietnam initially supporting the South by providing advisors and resources for nation and military building. After realizing that the government was corrupt, unpopular, and contemplating talks with the North, Kennedy supported its overthrow by a military junta. All this political turmoil 75 strengthened the North and destabilized the South (Holmes 2001, 954; Logevall 1999, xx). Eisenhower had established a U.S. military presence of 600 people, but between 1962 and 1963 Kennedy raised that number to 17,500. These were military trainers, advisors, observers, and combat support personnel. They were there to strengthen the South, not to fight a war. Kennedy had resisted a heavier commitment because of the poor government in the South, little support from allies, and little American public support. He would publicly say that it was the South?s war, and it was up to them to win it (Hess 2001, 78). At the time of his assassination, Kennedy was seeking other remedies such as improving relations with the USSR. Unfortunately, these plans were sidetracked when Kennedy was assassinated in November of 1963 (Kaiser 2000, 3-5). In 1964, some of the military personnel in Vietnam became casualties of the war. In May of 1964 the Viet Cong sunk the USS Card docked in Saigon, in August they attacked the USS Maddox, in November they attacked a U.S. base, and in December they bombed a hotel where American military were staying (Holmes 2001, 954). It was the administration?s handling of the attack on the USS Maddox that led to the fateful U.S. congressional decision to allow President Johnson unhampered freedom to escalate the number of military personnel in Vietnam (Holmes 2001, 954). Known at the ?Gulf of Tonkin? incident, three Northern torpedo boats attacked the U.S. destroyer Maddox on the 2nd of August, 1964. They did no damage except for a one inch bullet hole and were quickly run off. Two days latter, on the 4th of August, while the Maddox was on patrol with the U.S. destroyer Turner Joy, the Turner Joy experienced a false alarm indicating that there was another attack. Although almost no damage was done during the first attack and the second reported attack did not even occur, President 76 Johnson used these incidents as an opportunity to order a military air strike on northern naval bases and to get permission from Congress to escalate the war. The air strike was the first punitive, retaliatory strike of the war, and it resulted in the loss of ten percent of the north?s petroleum reserves (Gelb 1979, 101). This action led to a volley of strikes by both North Vietnam and the U.S. which resulted in Johnson?s execution of a U.S. Air Campaign called Rolling Thunder. Implementing a general military campaign rather than only responding to specific incidents is what many historians point to as Johnson?s fateful decision to escalate the war. After this point, there was no turning back (Holmes 2001, 954). This decision mired the U.S. in ten years of a war that was never officially declared, that most military experts today feel never could have been won, and which left over 58,000 U.S. soldiers dead and 153,000 wounded, as well as over a million Vietnamese dead (Summers 1999, 111-113). Lyndon Johnson and Vietnam When Johnson took office after Kennedy?s assassination, he showed little interest in the issue of Vietnam. It was only mentioned in passing in his 1964 State of the Union message. His two directives concerning this area were to reestablish Kennedy?s policy as his own and to authorize military planning for possible future operations. His priorities were his ?Great Society? (war on poverty) legislation and the next election (Gelb 1979, 97). Due to the directive allowing for military contingency planning and the instability of the government in the South (there were seven new leaders in 1964), everyone on Johnson?s staff had their own ideas about what to do about Vietnam. Most of Johnson?s advisors wanted to hold off taking military action but a few, including the Chairman of 77 the Joint Chiefs of Staff, were calling for air strikes (Gelb 1979, 100). Senator Mike Mansfield raised concerns that any air involvement could result in U.S. losses that would force an ill considered U.S. escalation. Johnson listened to the advice offered and then had William Bundy, Assistant Secretary of State, prepare a resolution giving Johnson carte blanche to commit forces in any way he chose to fight Communist forces in South East Asia (Hanyok 2000, 46). Johnson and his three primary advisors, Dean Rusk (Department of State), Robert McNamara (Secretary of Defense), and McGeorge Bundy (National Security Advisor and William Bundy?s brother) were known as the ?awesome foursome.? One would think that at least the Secretary of State would seek a diplomatic solution, but none of these advisors tried to dissuade Johnson from escalating in Vietnam. Johnson made it clear that he did not appreciate dissent. Anyone who disagreed with him lost access to him and lost influence on the policy they wanted to change. When Vice President Hubert Humphrey advised withdrawal while trying to point out how Vietnam was fundamentally different from Korea, he was no longer invited to meetings where discussions on Vietnam took place. When Canada wanted to stop the bombing, this ally was no longer included in the decision process (Logevall 1999, 394; Hess 2001, 92). Johnson remembered how talks with Hitler in Munich only encouraged the Nazis and convinced them that no country would stand up to them (Logevall 1999, 394). Janis believes that the ?awesome foursome? formed a group vulnerable to groupthink. He ascribes groupthink to dysfunctional elite decision-making groups that do not seek out or examine new alternative actions due to intense group cohesiveness. Once the group or group leader has hinted at what alternative he or she favors, the rest of the 78 group members quickly come on board, promoting and praising that alternative. No other alternatives are even suggested for fear of losing support of the group (Janis 1972, 8). Janis makes a good case that the escalation decision resulted from the groupthink syndrome. Johnson?s advisors were loyal and dedicated to him, and there appeared to be a great deal of hesitation in reevaluating past decisions or disagreeing with Johnson (Janis 1972, 101 ? 136). Even though George Ball, U.S. Under Secretary of State, usually played the devil?s advocate, everyone knew that it was part of his job and that he would be supportive of whatever the president decided. But Johnson was clearly in charge and making the decisions. Even though groupthink may have contributed to the mistake, game theory can still be used to evaluate it. There is a fine line between groupthink and authoritarianism, manifesting very similar symptoms. Dixon presents a theory that the worst military leaders are those who are the most authoritarian because they do not have the flexibility, creativity, and self confidence necessary to be good military leaders (Dixon 1976, 256-279). Dixon?s main point is that the traits needed to do well in a military environment are not the traits needed to be a successful military leader that can competently deal with the changing fortunes of war (Dixon 1976, 256-279). It is likely that the awesome foursome was influenced more by Johnson?s overbearing behavior then by groupthink. Rather than looking at group influences on Vietnam, Gelb looks at system influences, primarily the U.S. policy decision making system. Gelb does not think that this system broke down, but that it may have worked too well. He points out that the U.S.?s cold war goal was the containment of communism, and this goal was consistently 79 followed. He also points out that although most people wanted to end the war, the preference was for escalation rather than withdrawal and that there were no illusions that the war could easily be won (Gelb 1979, 2). Gelb agrues that the U.S. policy decision making system worked because it followed consistent goals, responded to public opinion, and realistically understood the alternatives and possible outcomes of those alternatives. Vietnam was a failure of foreign policy because of bad decisions made at the highest level and once made, the system took over to bring them to fruition (Gelb 1979, 366). Although Gelb does not address the decision-making models directly, his book seems to support the idea that OBM did not come into play until after Johnson made the decision to escalate. When Johnson found out that the Maddox had been attacked and had not destroyed its attackers, he wanted to know why a large, well equipped Navy could not win over a poor backward one. The reason was that the current rules of engagement did not allow pursuit. They only allowed ships to return fire but not follow (Gelb 1979, 101). Upon hearing of the false alarm on the Turner Joy, Johnson hid the fact that nothing happened and used the incident as an opportunity to ramrod the resolution drafted by William Bundy through Congress. It was approved August 7, 1964 and called the Gulf of Tonkin Resolution (Hanyok 2000, 46). The resolution gave Johnson the power he needed to escalate the war even though those voting for the resolution in Congress understood that Johnson would only use the authority to retaliate against attacks (Kaiser 2000, 337). Changes were not made to make the wording consistent with this understanding because the Democrats wanted to support Johnson in his election bid (Kaiser 2000, 337). 80 Johnson pushed his resolution for three reasons: to rebut the accusations of his opponent for president Barry Goldwater that he was not tough on communism, to give him flexibility with his actions, and to scare the North with his possible intention to escalate the war (Hess 2001, 88). But this resolution might have also restricted his perception of his options. When the North attacked a U.S. base in February 1965, Johnson executed an air campaign being planned by the military called Rolling Thunder which called for many more soldiers to be deployed to Vietnam (Hanyok 2000, 46). Soon after, to the surprise and dismay of his closest advisors, he added ground troops to fight the Viet Cong and shore up the government of the South (Hess 2001, 89-90). In September 1964 William Bundy sent a memorandum to Johnson stating that he had three alternatives in relation to Vietnam: pull out, limited support to the South, and full commitment to war. Bundy pointed out that Johnson was currently providing limited support as both Kennedy and Eisenhower had before him (Gelb 1979, 108). Robert McNamara, his Secretary of Defense, argued that a Communist Vietnam would have dire repercussions for Laos, Thailand, Cambodia, Malaysia, and Burma and that what the U.S. did in SE Asia would foreshadow the takeover of governments throughout the entire third world (Warner 2003, 830). Senator Richard Russell of Georgia wanted the U.S. to pull out and suggested that we instigate a coup in the South so that the new government would ask us to leave. He saw this as a way to disengage without losing credibility (Warner 2003, 832). The journalist Walter Lippmann met with Johnson to push for a diplomatic solution (Warner 2003, 833). Senate Majority Leader Mike Mansfield urged Johnson to pursue an international agreement neutralizing Vietnam, Cambodia, and Laos or else the U.S. would end up with a war in direct odds with China (Hess 2001, 83). George Ball, 81 the Under Secretary of State, felt that increased military action would only result in a stiffening of resolve by the North and that the U.S. should try to negotiate with the North without escalation and tell the South that the U.S. would pull out if the South did not develop a strong government supported by its citizens. He pointed out that the U.S. could end up like the French during the Indochina War, fighting a guerrilla war that could never be won. Unfortunately, Ball was viewed as having been assigned the role of ?devil?s advocate? and was taken less seriously for that reason (Warner 2003, 840). America?s allies were not in support of military action in Vietnam. France favored diplomacy, calling for a neutralization of the South (Kaiser 2000, 315). Japan called for a political solution for fear of getting caught in a war with China and unsure that the South could be held. Getting involved in another conflict in Vietnam seemed to be viewed as poor judgment by the international community (Kaiser 2000, 350; Logevall 1999, 380). In April 1965 seventeen unaligned nations approached the United Nations asking for immediate negotiations to end the conflict (Simons 1971, 182). Only four countries supported the administration: Australia, South Korea, Thailand and the Philippines; only Australia had no reservations (Logevall 1999, 378). Both internationally and nationally there was no shortage of voices counseling against U.S. military action in Vietnam. The intelligence community did not fully support the domino theory but pointed out that any country lost to communism would make China and the USSR stronger. Intelligence experts also felt that China would not join the war unless the U.S. bombed too deeply into Northern territory. Growing differences between China and the USSR could delay a Chinese response (Warner 2003, 839). The intelligence experts also 82 pointed out that the North would not give up easily and perhaps would never give up no matter what happened (Logevall 1999, xix). Johnson told McGeorge Bundy, his National Security Advisor, that he could not change the current policy until after the 1964 presidential election (Warner 2003, 831). Johnson seemed to agonize over the decision because he would tell Bundy that he saw no purpose in sending people to die in Vietnam, and that no one else was supporting the SEATO treaty so why should we, but in the same breath he would speculate that running from the Communists will embolden them further (Warner 2003, 833). This may have been a way to delay a decision until after the election. After the election, William Bundy again presented Johnson with three alternatives: status quo, quickly increasing the military tempo of attacks, or slowly increasing the military tempo of attacks in order to encourage the north to come to the negotiating table (Warner 2003, 839). Neither Johnson, nor his staff, ever seriously considered pulling out since it would be seen as a U.S. defeat; only different levels of commitment were argued. At that time it seemed important to stand firm against communism as the U.S. did with the Berlin Airlift, in Korea, and during the Cuban Missile Crisis. If the U.S. did not, then like the Nazis after Munich, Communist aggression would escalate (Hess 2001, 80). But if pulling out was never an option, then escalation was inevitable (Gelb 1979, 110-111). In fact, there was a fear of early negotiation, especially before the U.S. could demonstrate a military capability. The administration wanted to make sure that if a treaty was signed, it would stick, and the North would not turn around and overrun the South as soon as the U.S. pulled out (Logevall 1999, xx). 83 On December 3rd of 1964 Johnson approved the implementation of a two phased approach. The first phase, which would last 30 days, would include air strikes against logistical routes and a strengthening of the South Vietnamese government. The second phase would only go forward if the North did not come to the table and the South improved its government. It consisted of more air strikes and a higher military tempo (Warner 2003, 841). As the days moved on into December and January, the South got progressively more unstable and the North continued to make bomb attacks on U.S. troops. By February 1965 the North had killed another seven and wounded 109. An air strike was called in retaliation, which marked the beginning of Rolling Thunder, the first air campaign of the war (Warner 2003, 843). The administration stated that this escalation was to maintain a free South, but the administration did not contribute to this freedom when it secretly prevented the South from seeking some kind of accommodation with the North (Logevall 1999, 387). The purpose of this air campaign was to cut off Northern support to the insurgents in order for them to be neutralized by the South and to force the North to the negotiation table. In April of 1965, two months after the start of Rolling Thunder, Johnson paused the bombing and offered the carrot that he hoped to offer after his demonstration of force. It was a huge economic development package to benefit both the North and the South that the USSR would also participate in if the North would allow the South to stay independent. Johnson was confident that Ho Chi Minh would jump at the chance to end the hostilities and benefit his people. China and the USSR reacted negatively to this proposal, but more importantly, Minh clearly and completely turned Johnson down in his 84 national newspaper and also rejected attempts by the British to broker a peace (Hess 2001, 96; Simons 1971, 181-182) With this failure to find common ground between the U.S. and North Vietnam government, the military began asking for many more troops as they realized what would really be needed to execute such a war. The escalation would have actually been quicker had the infrastructure been in place to accept it (Kaiser 2000, 311). In June 1965 Johnson told McNamara that he could no longer see any way to win diplomatically or militarily and to pull out would be a humiliation for the U.S. (and for Johnson) (Hess 2001, 97). In July 1965 Johnson met with McNamara (Secretary of Defense), Clark Clifford (unofficial counselor) and a few others at Camp David. Clifford clearly supported pulling out of Vietnam. He advised Johnson that Vietnam was not important enough in the fight against communism for such a high military commitment, and he questioned the U.S.?s ability to win such a war. Clifford felt that staying would leave a worse impression on the international community than leaving. Clearly there were many influential people trying to convince Johnson not to escalate, but Johnson could not lose face in Vietnam because it would impact his ability to forward his Great Society programs. Although he did not ask for a declaration of war or mobilize the reserves or consider the use of nuclear weapons, Johnson took all other actions to Americanize the war in Vietnam (Hess 2001, 104). Rational Actor Model At the start of this particular game under study both players had picked limited war as the prior game alternative. At that point in time, Johnson was clearly aware of all the following alternatives: 85 U.S. Alternatives discussed at the time: Pull out militarily: either due to international neutralization, engineering a coup where the South asks the U.S. to pull out, or by claiming a weak South or some other similar reason Using diplomacy with the North, reach an agreement Continue limited military support while strengthening South Vietnam?s government and military Escalate military support while strengthening South Vietnam?s government and military The North has a similar set of alternatives: North Vietnam Alternatives: Pull out of the South to 17th parallel Using diplomacy with U.S., reach an agreement Continue limited military tempo Escalate military tempo A game matrix developed from these alternatives are shown in Table 5.1. Table 5.1: Johnson-Minh Game Theory Matrix Minh Pull Out Stay limited Negotiate Escalate Pull Out Stay limited Negotiate Johnson Escalate 86 The rules of the game as Johnson saw them were that, in the series of games called the Vietnam Conflict, communication between players was allowed in a formal diplomatic setting but agreements between players were not necessarily binding. Both players could make an unlimited number of moves any time they wanted to, although, for this particular game, Johnson chose first, then Minh, making it a consecutive game. It is unlikely that there would ever be a zero sum game, but the games would become negative sum if both players continued with military action (both losing people and money) and positive sum if both negotiated in good faith. From his discussions with his staff and others, Table 5.2 below describes how Johnson probably viewed the payoffs of different alternatives. Benefits are listed above the dotted line under each column, and costs are listed below the dotted line. 87 Table 5.2: Payoff of Different Options for Johnson over Vietnam Pulling Out Diplomacy Continued Limited Support Escalation and Winning Escalation and Losing U.S. lives will not be lost and money will not be spent Doves will be pleased ....................... Communist aggression will escalate U.S. will be blamed for every country which goes Communist U.S. will look weak, impotent and inconsistent Johnson will be responsible for the first U.S. defeat U.S. will be perceived as not honoring prior commitments (loss of credibility) Could lose ability to establish Great Society legacy Provides a way to militarily disengage without losing face U.S. lives will not be lost and money will not spent Doves will be pleased ....................... U.S. will appear weak Communist aggression will escalate like Nazi aggression after Munich Communists will not keep their word U.S. will be perceived as not honoring prior commitments (loss of credibility) Will not disrupt election process .................... ... U.S. lives will be lost and money will be spent Will convince allies of U.S. dependability by supporting obligations Will demonstrate U.S. will and determination Will prevent the lost of South Vietnam to communism and head off the loss of S.E. Asia and other third world countries Will prevent Chinese and/or USSR aggression and expansion Will not become the first U.S. president to lose a war Will win the next election Hawks will be pleased ...................... Will precipitate a war with China and/or USSR Will lose U.S. lives and money Will convince allies of U.S. dependability by supporting obligations Will demonstrate U.S. will and determination ....................... Will demonstrate administration? s poor judgment to U.S. allies Will lose the South and probably all of S.E. Asia to communism and Chinese expansion Will become the first U.S. president to lose a war Will lose the next election U.S. lives and money will be lost Will lose ability to establish Great Society legacy 88 The alternative with the worst payoffs would be escalation, whether the U.S. lost or won. Winning in Vietnam could result in a larger war with China or the USSR, which could be catastrophic if it developed into a major war where nuclear weapons were used. This was not an unprecedented scenario since only 13 years earlier General MacArthur was pushed out of North Korea by the Chinese when his forces got too close to the Chinese border. Losing in Vietnam could result in lose of lives, money, and countries to communism. The two alternatives with the best payoffs would be escalation and winning or diplomacy. Assuming that the North picks the alternative that gives the U.S. the worst payoff, then they would step up their military actions and not negotiate. A rational actor would always hope for the least amount of effort for the best result. Both players would prefer the other player to select payoffs in the following order: pull out, then negotiate, then wage limited war, and finally escalate. Based on these preferences, a prioritization of outcomes or results can be developed. For example, if Johnson was making a decision, purely as a rational actor, his best outcome would be for both the U.S. and the North to stop military action (pull out ? pull out). This would be the least amount of effort for the best result and would therefore be Johnson?s number one priority. The same would be true for Minh. Their lowest priority would be increased military action on both sides (escalate ? escalate). Basing the matrix values on this prioritization, each player, if they were rational actors, would probably prioritize like Table 5.3 below. 89 Table 5.3: RAM Prioritization for Johnson-Minh Johnson Priority Result (Johnson-Minh) Minh Priority 1 Pull out-Pull out 1 3 Stay limited-Pull out 12 2 Negotiate-Pull out 6 4 Escalate-Pull out 14 12 Pull out-Stay limited 3 9 Stay limited-Stay limited 9 10 Negotiate-Stay limited 7 11 Escalate-Stay limited 15 6 Pull out-Negotiate 2 7 Stay limited-Negotiate 10 5 Negotiate-Negotiate 5 8 Escalate-Negotiate 13 14 Pull out-Escalate 4 15 Stay limited-Escalate 11 13 Negotiate-Escalate 8 16 Escalate-Escalate 16 Based on the above prioritization the RAM matrix payoffs could therefore look like Table 5.4 below. Table 5.4: Johnson-Minh Game Theory Matrix under RAM Minh Pull Out Stay limited Negotiate Escalate Pull Out 1,1 12,3 6,2 14,4 Stay limited 3,12 9,9 7,10 15,11 Negotiate 2,6 10,7 5,5 13,8 Johnson Escalate 4,14 11,15 8,13 16,16 Under the RAM, Johnson should assume that Minh would choose escalation to give Johnson his worst case scenario. Then, Johnson should chose the negotiate alternative to give him his best worst case scenario. The problem is that Johnson could 90 not negotiate unless Minh also picked negotiate. If negotiate-escalate is not a real outcome, then Johnson?s most rational choice is to pull out when he determines Minh would not negotiate since 14 is his highest priority under escalate. Johnson actually picked two alternatives: escalation (to demonstrate strength) and then diplomacy. In this game Johnson was trying to tell Minh, agree to my terms or face the wrath of the U.S. military. Johnson set up this game with Minh to force Minh to let South Korea stay independent. Ho Chi Minh had two alternatives, agree to Johnson?s terms or not agree. Johnson?s game would result in one of his best results (#5: negotiate- negotiate) or one of his worst results (#10 or #13). His best outcome was to keep the South democratic and uninfluenced by the USSR and China while slowly strengthening the South?s government and popularity and at the same time disengaging militarily. His worst outcome was to put himself in a position where he no longer had alternatives except to escalate. Once Rolling Thunder had begun, Johnson?s self esteem was on the line. If the North would not agree to his offer, he had to back up his threat of military force or look weak and naive in the eyes of the international community and the American public. Johnson set up a game that was not rational since it gave him the possibility of getting his worst possible outcome. Determining a Different Utility Function Since Johnson?s Vietnam decision did not follow the classic RAM, he must have either made his decision using a different utility function or he made a purely emotional decision that bore little resemblance to well considered rational thought. The first step in determining what his decision was based on is to look at what some authors of in-depth psychological studies believe were Johnson?s goals, objectives, and driving forces. All 91 look at Johnson except George and George who omits Johnson from their book without explanation (George and George, 1998). Lammers and Genovese (2000) paint a picture of a man who was ambitious and politically motivated. This is probably to be expected since Johnson worked in various positions for 30 years on Capital Hill, the last being the Senate majority leader, before becoming vice president. Johnson felt that as president, he should represent all interests. Johnson was also fixated by quantity rather than quality. The more policy he got passed the better job he thought he was doing. This desire for quantity rather than quality was also seen during the Vietnam War when the enemy body count because the measure of success. He never seemed to stand back and look at what he was accomplishing and where he was going. He tried to seek consensus and with that consensus, admiration. (Lammers and Genovese 2000, 94-95) From this analysis, it would seem that Johnson most followed the Governmental Politics Model (GPM) where decisions are made through bargaining and compromise and his primary goal was to keep constituents and interest groups happy in order to stay in power (Allison and Zelihow 1999, 255-313). Greenstein (2004) also sees Johnson as a political animal. Johnson was an expert at bullying others to his side and would get legislation passed with only a small margin of votes. Yet he was severely lacking in vision (Greenstein 2004, 86-89; Caro 2002, 581- 596). He wanted the Vietnam issue to go away but stumbled into more and more military force with no idea of the final outcome he wanted to attain. Even with his Great Society policies, he never developed a comprehensive plan that could significantly improve the lives of the poor. Although he could remember copious amounts of information, he could not organize it into actions that would accomplish specific goals. Meetings were not held 92 to objectively discuss alternatives but to figure out how to silence critics and make the administration look good (Greenstein 2004, 86-89). Again, the GPM emerges as the best model to describe Johnson--politically motivated, trying to present the best image possible before the next election, and not thinking about the wider implications of his decisions. According the Barber (1992), Johnson?s key characteristic was his style of getting things done through one-on-one contacts with people. If he needed the help of someone who did not work for him, he would be flattering, helpful, and persuasive. But the people who worked for him were bullied and dominated (Barber 1992, 66-68). Barber posits that Johnson had an active-negative personality which means that he expended a great deal of energy on accomplishing political functions and sought opportunities that would result in an increase in his political power. Barber?s (1992) characterization of his four personality types places Johnson as an active-negative. Johnson would work far more than is necessary and would feel like a martyr for doing so while seeking continual positive reinforcement from others. He also would see himself at restraining his aggression which compels him to ?fight or quit? and focus this aggression on a personal enemy. According to Barber, the active-negative sees people as weak or grasping and as the primary threat. He consistently uses the same political style to manage all situations. Also, once a course is laid out the active-negative will follow it to the bitter end, not considering other alternatives (Barber 1992, 80-83). In Barber?s description, the GPM still stands out since Barber states that Johnson handled all situations politically although there may be a little of Elitist Actor Model (EAM) also in his view of people as weak or grasping. 93 Governmental Politics Model (GPM) In the GPM decisions are made through bargaining and compromise. Goals and objectives are not consistent but depend on the strength of specific interest groups, public opinion, and constituency ideology. The rules of the game are the political rules first and foremost, which means that fund raising and keeping constituents happy are keys to staying in power (Allison and Zelihow 1999, 255-313). If Johnson?s primary objective was to please constituents, interest groups, and party members in order to stay in power, he would not necessarily seek to avoid the worst outcome for the U.S., he would primarily seek to avoid the worst outcome for Lyndon Johnson. The most embarrassing immediate outcome would be orchestrating the first U.S. defeat and making the U.S. look weak and impotent, and by extension making Johnson look weak and impotent. From a GPM viewpoint, the best decision would be to keep the war going at a limited level, committing the least amount of resources possible, but enough to encourage the North to negotiate for peace. The problem with such an alternative is that the North Vietnam actor chose escalation against the U.S. If the North had only acted against the South, the limited support alternative would have worked, but the North kept targeting U.S. personnel, and since Johnson already had congressional support to escalate, he had painted himself into a corner once the North rejected his overture for talks. Even if Johnson had secured the Gulf of Tonkin Resolution to keep his options open, he actually left himself only one alternative. Every U.S. life that was lost brought more and more pressure on him to win the war. Johnson did not want the country or his 94 party, and by extension himself, to appear weak in the face of an enemy. He made it clear that he would not be the first president to lose a war (Logevall 1999, 392). The GPM emphasizes bargaining and compromise. There were many people against the war, and it was in Johnson?s best interest to get in and out as quickly as possible. Since he knew he could not do this militarily, he tried to force a negotiation by providing a show of force as a threat and then a generous economic package as an incentive. In Johnson?s mind, no politician could refuse such an offer. Johnson did not want to lose support by appearing weak. He miscalculated Ho Chi Minh?s response to his offer for negotiation. Johnson did not consider that the other player (Mihn) would pick the option that would give him his worst alternative. Johnson set up a game that committed the U.S. to a long and lengthy war if Minh refused to negotiate. When he would not give him the one thing he had to have to win, a non-Communist South Vietnam, Johnson could not back away. He had not left himself any alternatives except to slowly drag the U.S. into a protracted war and win the honor of being the president who, based on the McConnell Center survey, made the third worst mistake in the history of this country. Using the GPM does not change Johnson?s priorities. He still wanted the other player to pull out the most, than negotiate, wage limited war, and least of all escalate military action. The problem is that GPM takes ?pull out? off the table as a viable alternative in Johnson?s mind. GPM did not change the priorities. What it did was remove an alternative. If pull out was not a viable alternative under GPM, then RAM works as shown by Table 5.5 below. 95 Table 5.5: Johnson-Minh Game Theory Matrix under GPM Minh Pull Out Stay limited Negotiate Escalate Stay limited 2 , 8 7 , 5 5 , 6 11 , 7 Negotiate 1 , 2 8 , 3 4 , 1 10 , 4 Johnson Escalate 3 , 10 9 , 11 6 , 9 12 , 12 If GPM removes the pull out alterative, then the best worst alternative is negotiate, which as pointed out earlier, can not be done unless both sides pick the negotiation alternative. A rational U.S. actor assumes that the North will step up its military actions, so picking the diplomacy alternative would be rational until it is determined that the North will not negotiate. Picking continued limited support would then be the best alternative, but in the real world the North?s increased tempo would inevitably precipitate increased tempo of the U.S. actor. Johnson did not initially pick limited action though; he picked escalation in order to make a show of strength prior to negotiation. Picking escalation would bring the possibility of a worst case scenario. Even though, under GPM, escalation was inevitable, Johnson?s actions did not demonstrate a rational process which would have been picking negotiation, then limited military response. Conclusion In this case study, use of the RAM does not explain Johnson?s decision to escalate the Vietnam War, but using the GPM objectives, the decision makes much more sense but still does not fully inform a utility function for the RAM. 96 CHAPTER 6 RICHARD NIXON: WATERGATE Richard Nixon was vice president during Dwight D. Eisenhower?s term of office from 1953 until 1961. He ran against John F. Kennedy for president in 1960 and then for California governor in 1962, losing both races. In spite of these losses, Nixon became the Republican nominee for president against Hubert Humphrey in 1968. He won based on claims to represent the ?silent majority? and on a secret plan to end the war in Vietnam. In office he seemed particularly talented in the area of foreign policy (Kutler 1990, 60; Nixon 1978). Nixon could have had a successful and prestigious administration, reaching arms agreements with the USSR, opening relations with China, and ending the Vietnam War. Instead Richard Nixon?s presidency ended in shame, as he resigned in order to avoid impeachment and possibly jail (Emery 1994, xi; Kutler 1990, 616-617; Olson 2003, 183). Historical Context Richard Nixon won a House seat in 1946 as a Republican from California and a Senate seat in 1950. In both cases he used fear of communism to his advantage, by suggesting that his opponents supported organizations sympathetic to communism. In 1952 he ran as Dwight D. Eisenhower?s running mate but was almost removed from the ticket when a business fund was discovered that Nixon used for personal expenses. He 97 successfully defended himself and remained on the ticket becoming vice president (Haldeman 1978, 80; Kutler 1990, 49-50; Nixon 1978, 31-115). In 1960, Nixon ran against Senator John F. Kennedy for the presidency. The race was very close but Nixon probably lost due to Eisenhower?s lack of support and Kennedy?s sterling performance during the televised debates. So in the end, Kennedy became president. Nixon also lost the 1962 California Governor?s election after which he became a senior partner in a New York City law firm and began to quietly rebuild his Republican base (Ehrlichman 1982, 11; Kutler 1990, 54; Nixon 1978, 229-295). In 1968, Nixon ran for president against Hubert H. Humphrey, the Democratic candidate, and George Wallace running as an Independent. This was during a time of great social upheaval in America. Student demonstrations and a hippie counter culture had developed as a reaction to the war in Vietnam and were fueled by government deceit on how the war was progressing (Kutler 1990, 97; Magruder 1974, 113; Nixon 1978, 295-362). The hippies believed in ending the war and uninhibited sexual relations (make love not war). This was toward the end of the African-American Civil Rights Movement and at the beginning of the Black Power Movement. Race riots occurred in several major cities from 1963 to 1970. Nixon ran as the candidate who represented the majority of Americans (Ehrlichman 1982, 203; Nixon 1978, 295-362; Olson 2003, 182). He termed these Americans the ?silent majority? who were opposed to hippies and other anti-war demonstrators (Emery 1994, 14; Kutler 1990, 192; Magruder 1974, 98; Nixon 1978, 414). Nixon also promised an honorable peace in Vietnam, although he never disclosed how he intended to accomplish such a feat. This platform resonated with voters and he 98 went on to win the election against Humphrey and Wallace to become the 37th president (Nixon 1978, 298; Olson 2003, 14). As president, Nixon was prolific in his actions concerning both foreign and domestic policy. In Vietnam he endeavored to turn the war over to the South Vietnamese military by providing weapons and training while reducing American troops. He reinitiated an Air Force bombing campaign of North Vietnam called Linebacker, as well as secret bombings of Cambodia and Laos, in order to encourage the North to agree to release American prisoners of war and cease attacks on the South. The bombings of Cambodia and Laos were kept secret since they were neutral countries, but Nixon thought that it was necessary to destroy supply routes from the North (Emery 1994, 17; Kutler 1990, 157; Magruder 1974, 109; Nixon 1978, 381). Unlike Johnson?s Rolling Thunder bombing campaign, Linebacker brought the North to the table for several reasons. Perhaps most important was the fact that Nixon had accomplished a higher level of rapport with China and the Soviet Union, making their entry into the war much more unlikely and allowing him to send bombers further north than Johnson dared go. Also, the North had now come to understand that U.S. public opinion was against the war and Americans did not want Vietnam as a colony the way the French had (Emery 1994, 14; Kutler 1990, 604; Nixon 1978, 704). When China and the Soviet Union began to pull away from each other, Nixon took the opportunity to visit China and open a dialog with her leaders. As visits and cultural exchanges with China began to take place, the Soviet Union, fearing a U.S.- Chinese alliance, also accepted invitations to talk. In the end, China won a seat in the United Nations and the U.S. and Soviet Union signed a treaty (SALT I) to reduce nuclear 99 weapons and both China and the USSR encouraged North Vietnam to reach a peace agreement with the U.S. (Emery 1994, 4-5; Kutler 1990, 604; Nixon 1978, 524). On the home front, Nixon supported the recent Supreme Court decisions requiring racial integration of schools by setting up local committees to reorganize school districts for maximum integration with minimum busing (Nixon 1978, 435-445). He also supported the first minority affirmative action program called the Philadelphia Plan which forced government contractors to change their hiring practices (Lammers and Genovese 2000, 246). Nixon?s first term was successful enough to allow him to be elected to a second term over George McGovern, whom he depicted as a leftist (communist leaning) extremist, with the greatest landslide victory of any president in history, losing only one state and wining over 60 percent of the popular vote (Kutler 1990, 237; Nixon 1978, 716). Richard Nixon and Watergate During this time, J. Edgar Hoover was director of the Federal Bureau of Investigation (FBI) and had been director since 1924. Hoover had negative files on many politicians, and he did not like cooperating with other intelligence gathering agencies or sharing his files with anyone, including the White House, which greatly bothered Nixon (Colodny 1991, 98; Emery 1994, 11) Kutler 1990, 97; Nixon 1978, 472). The invasion of Cambodia in May 1970 and the invasion of Laos the following year touched off huge antiwar protests. The deaths of four Kent State students at the hands of the Ohio National Guard further fueled the flames of dissent (Colodny 1991, 103; Emery 1994, 18; Kutler 1990, 155; Magruder 1974, 101; Nixon 1978, 457). 100 Nixon?s reaction was to seek personal information on left wing groups and other political enemies. He wanted to gather anything he could later use to secure their support or to smear their reputation. Nixon wanted files similar to those kept by J. Edgar Hoover, but the kind of information he wanted could not be legally gathered by the Central Intelligence Agency (CIA) and FBI (Colodny 1991, 99; Emery 1994, 5). Instead he employed several private operatives and made it known that the way to get ahead in the administration was to collect useful intelligence on his enemies (Colodny 1991, 95; Kutler 1990, 112; Magruder 1974, 166; Nixon 1978, 513). Many of Nixon?s staff, especially on his reelection committee, played on Nixon?s pettiness and paranoia to gain power at the White House. They developed a list of political enemies, along with methods by which to hurt these enemies, such as having their taxes audited (Colodny 1991, 104; Kutler 1990, 108; Magruder 1974, 167). In June 1971, the New York Times began to publish documents that were part of a secret Department of Defense (DOD) study on Vietnam. This document became known as the Pentagon Papers. The document had much more to do with the Kennedy and Johnson administrations than with Nixon, but Kissinger convinced Nixon that this was a great breach of national security, perhaps because Kissinger had been involved with the two prior administrations (Colodny 1991, 112; Emery 1994, 39-40; Kutler 1990, 110; Nixon 1978, 508). Although the DOD, FBI, and Department of Justice were all investigating the leak, Bud Krogh, a White House aide, was also asked to look into it. Krogh, along with David Young, a former aide to Kissinger, formed a task group called the ?plumbers,? whose job was to investigate leaks such as the Pentagon Papers. Soon there was a new leak to the New York Times of a classified negotiation strategy for the 101 on-going Strategic Arms Limitations Talks (SALT) with the USSR (Colodny 1991, 113; Emery 1994, 53; Kutler 1990, 115; Magruder 1974, 171). The new leaks placed pressure on the plumbers to take action. Their first undertaking was to break into the office of Dr. Lewis Fielding, the psychiatrist of Daniel Ellsberg, the civil servant who leaked the Pentagon Papers, in an effort to find information that could discredit Ellsberg. Although John Ehrlichman, the president?s Assistant for Domestic Affairs knew that Fielding was being targeted for investigation, he may not have been aware of the details, and if so, Nixon may not have been aware either (Colodny 1991, 114; Emery 1994, 69; Ehrlichman 1982, 367; Liddy 1980, 231; Magruder 1974, 171; Nixon 1978, 514). Nixon did not want to use the infrastructure of the Republican Party to organize has reelection campaign which is the mechanism candidates who become the party nominee normally use to run a presidential campaign. Instead, Nixon formed a new organization called the Committee to Re-elect the President (CRP), pronounced CREEP. This allowed Nixon to control all aspects of his campaign by setting up the organization and naming personnel he preferred. One key aspect of the CRP was the intelligence gathering arm. John Mitchell, CRP Director, and H.R. Haldeman, White House Chief of Staff, primarily asked for information on the Democrats and on the anti-war movement (Colodny 1991, 115; Haldeman 1978, 34; Magruder 1974, 146; Nixon 1978, 542). G. Gordon Liddy, who was originally a plumber, became general counsel for the CRP. In his duties as counsel he employed several people whose job it was to try to detect which government officials were leaking information to the press and to infiltrate the headquarters of Nixon?s possible opponent, Senator Edmund Muskie, and to discredit 102 anti-war demonstrators (Emery 1994, 55-57; Haldeman 1978, 40; Olson 2003, 37). In order to gather information on Senator Muskie, Liddy decided he needed to bug the Democratic National Committee (DNC) which was located in the Watergate complex in Washington, D.C. In general, Liddy would determine how to collect the intelligence information requested, but in the case of the Watergate break-ins Liddy got approval from Jeb Stuart Magruder, CRP Deputy Director, who in turn had gotten approval from John Mitchell to bug the Democratic Headquarters at the Watergate Complex (Colodny 1991, 134; Emery 1994, 53; Liddy 1980, 303; Magruder 1974, 195; Nixon 1978, 635). There was never any concern about an operation being illegal, only whether it would provide the desired information, whether it was worth the time and money and whether the perpetrators were likely to be caught (Magruder 1974, 195) On Sunday evening May 28, 1972 a small group of plumbers headed by Liddy and directed by Hunt broke into the DNC headquarters in the Watergate Office Building where bugs were placed on phones, photos were taken and desks searched (Colodny 1991, 136; Emery 1994, 121; Olson 2003, 38). The logs of the conversations overheard from the bugged phones went from Liddy to Magruder to John Mitchell. But there was little in these conversations that was useful (Colodny 1991, 147; Liddy 1980, 321). Liddy planned a second break-in of the DNC with the specific objective of finding any derogatory information the Democrats might possess that they could use against Nixon and to fix an inoperable bug in Larry O?Brien?s (Chairman of the DNC) office (Colodny 1991, 148; Liddy 1980, 325). On the evening of June 16, 1972 McCord entered the DNC pretending to deliver a typewriter, but on his way out he used tape and paper to keep doors from automatically closing. The paper and tape were noticed by the 103 security guard on his midnight rounds; so, he re-secured the doors thinking maintenance crews were responsible. When the plumbers, headed by McCord, found the door locked, they picked the lock and re-taped it open (Colodny 1991, 155; Emery 1994, 132; Liddy 1980, 335). When the security guard again found the door taped, he called the Washington, D.C. police. McCord and four other plumbers were apprehended by the police when they were found hiding behind desks during the early morning of June 17th (Colodny 1991, 157; Liddy 1980, 336; Magruder 1974, 217). The keys held by the burglars led police to their command center in a hotel room across the street. There the police found Liddy and Hunt?s finger prints (Colodny 1991, 160; Kutler 1990, 189; Liddy 1980, 337). The next day Magruder told Liddy to try to get McCord out of jail by talking to the Attorney General Dick Kleindienst since McCord would have provided a direct link back to the White House. Kleindienst declined to help. This was the first attempt to cover up the break-in, but this attempt seemed to not be linked back to Nixon. Sources disagree on whether Magruder was told to have Liddy contact Kleindienst, but if someone directed Magruder, it was probably at the Haldeman, Mitchell, or John Dean (White House Counsel) level and did not come all the way from Nixon (Colodny 1991, 167; Emery 1994, 145; Kutler 1990, 211; Liddy 1980, 345; Magruder 1974, 215). At the time of the break-in Nixon was vacationing in Florida. When Haldeman was notified of the break-in, he was confused about why such a break-in would occur since Nixon thought it was doubtful that any useful information could be gotten from the DNC (Colodny 1991, 173; Haldeman 1978, 27; Nixon 1978, 627). Magruder told Haldeman that the burglars were not technically working for CRP 104 since McCord owned a private security agency (Colodny 1991, 174; Haldeman 1978, 34; Magruder 1974, 220). Based on Magruder?s assurances Haldeman approved a press release which claimed that the CRP had no connection to the break-in. Again, it is debatable whether Haldeman suspected the press release was false, but he still approved it, and it became the second cover up attempt (Haldeman 1978, 35). Haldeman then called Ehrlichman, who checked with Chuck Colson, White House Special Counsel, to see if Hunt was connected with the White House. Colson told Ehrlichman that Hunt had been let go months before (Colodny 1991, 174; Haldeman 1978, 36). Haldeman talked to Nixon, who seemed unconcerned, saying that the voters will view it as a political prank, but Nixon was worried enough to talk to Colson personally about the situation (Emery 1994, 154; Haldeman 1978, 37; Nixon 1978, 630). At this point Nixon could have called for an internal investigation since he had not ordered the break-in and thought that none of his inner circle was involved, but the election was looming and connecting such an incident to the CRP could jeopardize his re-election (Colodny 1991, 175; Olson 2003, 175). The press quickly connected McCord to the CRP and also linked Hunt to the burglary since Hunt?s name and phone number were on the person of one of the burglars and a check he had signed was being held by another (Haldeman 1978, 35; Liddy 1980, 355; Nixon 1978, 628). Hunt was advised to flee the country providing a third cover-up attempt, although this decision was rescinded and Hunt was asked to stay (Colodny 1991, 181; Liddy 1980, 357). 105 High officials at both the White House (Haldeman, Ehrlichman, and Dean) and the CRP (Mitchell) were concerned about what was really happening. Other than Hunt, the burglars were from the CRP side. During this time, both Liddy and Magruder destroyed many CRP files to further cover up what had transpired (Colodny 1991, 185; Haldeman 1978, 40; Liddy 1980, 341; Magruder 1974, 227). Later Liddy was fired in an attempt to make him a scapegoat by saying that all illegal activities had been planned and executed by him without any prior knowledge of CRP (Magruder 1974, 234; Nixon 1978, 644). A story in the Washington Post the morning after the break-in tying Hunt firmly to the White House propelled Nixon to meet with his Chief of Staff on the subject (Emery 1994, 148; Colodny 1991, 186). Haldeman briefed Nixon on what he knew. Nixon did not want Hunt linked to the White House and tried to come up with explanations for the link between Hunt and Colson (Colodny 1991, 187; Nixon 1978, 634). There was never any serious discussion of having an internal investigation and coming clean with the press and the public or cooperating with the FBI investigation even though it appeared that Nixon was not directly connected with any of the burglaries prior to that point in time (Ehrlichman 1982, 312; Nixon 1978, 631). There were several reasons not to cooperate with the FBI. Nixon had a bank in Mexico where he was stashing political donations that he did not want the FBI to find because the donors wanted to remain anonymous (Emery 1994, 6; Haldeman 1978, 63; Nixon 1978, 638). Another reason for not cooperating was because, prior to the break-in, Nixon asked his people to look for damaging information on Larry O?Brien, Chairman of the DNC. In order to comply with the president?s request, O?Brien?s phone in the DNC 106 was tapped during the first break-in (Haldeman 1978, 209; Liddy 1980, 325; Magruder 1974, 180). If Nixon believed that his order to get derogatory information on O?Brien was the catalyst for the break-in or if he had actually known about and approved the break-in ahead of time, then he would certainly be reluctant to open himself up to an investigation prior to the election (Kuter 1990, 204; Nixon 1978, 514). Nixon?s conversation with Haldeman on June 20th was in Nixon?s office, and therefore taped, but 18 ? minutes were erased from that tape at a later date. Both Nixon and Haldeman?s memoirs agree on the gist of the discussion which was how to make the break-in appear to be unconnected to the White House (Colodny 1991, 187; Haldeman 1978, 42; Nixon 1978, 632-634). Nixon had found out from Colson that Hunt had a Bay of Pigs connection, and since most of the burglars were Cuban, he thought it would make sense to make it seem that the Cubans were trying to find out the Democrats? plan to handle the Castro issue. If the Cuban population in Miami could be convinced that the Democrats were supportive of Castro, the burglary would actually help Nixon?s reelection bid (Colodny 1991, 187; Ehrlichman 1982, 318; Nixon 1978, 634). Nixon later proposed setting up an ?anti-Castro fund? in order to collect money to pay the legal bills of those in jail and to aggravate Cuban opposition to McGovern, Nixon?s opponent (Emery 1994, 178; Haldeman 1978, 51; Nixon 1978, 635). In the mean time Nixon?s senior staff was trying to find out what had really happened and what they should do about it, but they wanted to keep this knowledge away from Nixon until after the election for fear that Nixon would do something to damage his campaign. Nixon, for his part, was never reported to have asked for the details behind the burglaries (Colodny 1991, 188). After Nixon had assured himself that everyone from 107 Mitchell on up had no prior knowledge of the incident, he treated it as merely a public relations problem that had to be dealt with until the media moved on to other things (Kutler 1990, 190; Nixon 1978, 635; Olson 2003, 177). The Democrats quickly filed a $1million civil suit in response to the break-in and the FBI started its own investigation (Nixon 1978, 629 and 636). Nixon was legally obligated to pass on to the FBI any information he had on the break-in, but he did not. At a press conference held five days after the break-in, Nixon denied all involvement (Colodny 1991, 193; Ehrlichman 1982, 323; Kutler 1990, 191; Magruder 1974, 230; Nixon 1978, 638). The election was coming up in less than six months. Even if Nixon was not specifically aware of the break-in before hand, he knew an investigation would uncover other things such as the names of campaign contributors that would not place him in the best light in front of the American public (Haldeman 1978, 60; Nixon 1978, 629; Olson 2003, 175). Also, he did not think he was doing anything the Democrats had not done. Nixon was certain that both Kennedy and Johnson had extensively used wiretapping and moles to gain intelligence information on opponents (Emery 1994, 5; Haldeman 1978, 27; Ehrlichman 1982, 32; Nixon 1978, 628). If Nixon had found out what really happened and had told the press and the public, he could probably have avoided the FBI investigation, but it would have still put him in a bad light prior to the election. It would have given the Democrats something to use against him. The handling of the cover up was disorganized and reeked of desperation. Nixon?s systematic lying and adherence to divergent cover stories demonstrated very sloppy and sinister thinking from a normally intelligent and well organized individual. 108 On one side he wanted Liddy to take full responsibility for the break-ins -- projecting him as a loyal zealot who had read too many spy novels while, at the same time linking the break-in to a CIA anti-Cuban operation (Nixon 1978, 633-645; Olson 2003, 180). The ?smoking gun? and point of no return was when Nixon agreed to let Haldeman call the CIA and ask the agency to claim that the break-in was a CIA operation in order to get the FBI to back off. Although the CIA was not supposed to operate domestically, it often did, and there was a standing agreement that the FBI and CIA organizations would not interfere with each other. The conversation with the CIA was recorded on tape and was the ultimate proof of a cover up (Colodny 1991, 201; Ehrlichman 1982, 315; Haldeman 1978, 62; Kutler 1990, 219; Nixon 1978, 641). This action dovetailed with Nixon?s earlier recommendation that the break-in was a Cuban operation connected to the Bay of Pigs, which was a CIA operation. Nixon advised Haldeman to threaten the CIA by telling its director that the FBI investigation had to be stopped or the break-in would be linked to the Bay of Pigs, making the CIA look bad (Colodny 1991, 201; Ehrlichman 1982, 325; Haldeman 1978, 62; Nixon 178, 642). The request that the CIA block the FBI probe was the final cover up action that sealed Nixon?s fate. Although it succeeded in containing the damage of the break-in until after the election, two years later, the taped conversation where Nixon agreed to use the CIA to stop the FBI, would end his presidency (Colodny 1991, 204; Haldeman 1978, 68; Ehrlichman 1982, 325). With the help of the Assistant FBI Director, Mark Felt, known at the time as ?deep throat,? word would get out to reporters who would keep the story before the public until Nixon found himself with no alternative but to resign. 109 Had Nixon not pursued the CIA part of the cover-up and had Liddy confessed and taken total blame, then the worst case legal scenario would have been that the CRP would be implicated, but the White House would have probably remained untouched. (Olson 2003, 178). The problem was that the worst case legal scenario could have had a devastating effect on the election. If the CRP were dirty, by implication, the voters might also view the White House as dirty, and the election could have gone to McGovern. Nixon had avoided the political worst case scenario while tumbling head first into the legal one. Rational Actor Model Douglas Muzzio?s Watergate Games, Strategies, Choices, Outcomes (1982), which examines key decisions surrounding Watergate, seems to be the only book that provides an in-depth analysis of a president?s political decisions using game theory. For every decision Muzzio prioritizes the alternatives for each player based on what the other player also does. So, for two alternatives there are four possible outcomes. These outcomes are then prioritized. Although Muzzio?s book does not deal with Nixon?s initial cover up decision, it is summarized here to give the reader an understanding of the context of the times, the likely thinking of the participants, and the game theory logic of another political scientist. First, Muzzio uses the game of prisoner?s dilemma to determine why James McCord and Howard Hunt did not make a deal with prosecutors before they were indicted. Muzzio justified their actions this way: If Hunt and McCord kept silent, the Watergate cover up would not be revealed, and they would be taken care of by Nixon. If only one player made a deal with the prosecution, then that one would get a reduced 110 sentence but would lose presidential support, and the Watergate cover up would become known. If both cooperated with the prosecution, then they would receive little reduced sentence, they would lose presidential support (money and a pardon after a year of jail as promised by John Dean), and the cover up would be out of the bag. So each player?s order of preference was to both be silent, for himself alone to make a deal, to both make a deal, and for the other to make a deal alone. Therefore, mutual silence became the Pareto optimal outcome which gives each player the best outcome and does not harm the other player. It is also the outcome at equilibrium since any change will result in a worse outcome for both. (Muzzio 1982, 20) The game changed for McCord when all the players were given long sentences unless they cooperated with the grand jury and Ervin Committee that were looking into Watergate. McCord did not want to be in jail so his preferences changed priorities to him alone making a deal, both making a deal, both keeping silent, and the other person alone making a deal. McCord was a family man with strong ties to his family, church, and community while Hunt had a spy mentality where keeping silent was always the best route. When it looked like McCord would testify to reduce his sentence, there was no longer any rational reason for Hunt to remain silent, so he testified also. (Muzzio 1882, 31-33) Until McCord agreed to testify, John Dean and Jeb Magruder were playing a similar game with the same preferences as the initial McCord-Hunt game until McCord made his deal with prosecutors. At this point the next higher level of conspirators was in danger, so there was no longer the conviction that the president would provide money and an early pardon since the blame was moving up the chain closer to the president. The 111 belief now was that the president was looking for someone to take the blame and that the first major player to cut a deal would get the most immunity and avoid getting blamed for the cover up. Also, Assistant U.S. Attorney Earl Silbert tried very hard to make John Dean think that Gordon Liddy had cut a deal and was talking to the grand jury about the cover up. So, the Dean and Magruder order of preferences was to cut a deal first, both cut a deal, both stay quiet, and the other cut a deal first. (Muzzio 1982, 39) In both games the initial best strategy was to be silent, but when things started unraveling and the president started looking for scapegoats, the game payoffs changed to cutting a deal as quickly as possible because Dean and Magruder expected the other player to cut a deal. Both cut a deal, but Dean did it ten days sooner than Magruder. (Muzzio 1982, 43) The key evidence concerning the Watergate cover up was the tapes that President Nixon made in the White House. Special Prosecutor Archibald Cox subpoenaed the tapes. Nixon appealed the subpoena to the District of Columbia Court of Appeals, basing his appeal on executive privilege, and lost. He did not appeal it to the Supreme Court for fear of losing. As a compromise, Nixon offered to provide narrative summaries, rather than authenticated verbatim transcripts, as long as future requests for information were prohibited. According to Muzzio, in the mind of Nixon and his staff, Cox had three alternatives: accept the offer, reject the offer, or reject and resign. If Cox resigned, Elliot Richardson, the attorney general who appointed Cox, would then have two choices: to stay and appear to support Nixon or to resign and appear to support Cox. Nixon believed that Richardson?s dominant strategy was to stay since he was a team player with lots of ambition. Nixon thought that if Cox and Richardson followed their dominant strategy, 112 then Cox would reject the compromise and resign while Richardson would remain. This would get Cox out of Nixon?s hair while at the same time Richardson, by not quitting, would appear to support Nixon over Cox. (Muzzio 1982, 93) Nixon was wrong in his evaluation of Cox. Cox did not have a dominant strategy to resign. Cox?s dominant strategy was to stay and fight, the worst outcome for Nixon. When Nixon told Alexander Haig to tell Richardson to fire Cox, Richardson refused and promptly resigned (a very different outcome than what Nixon was counting on). The same thing happened with the new attorney general, William Ruckelshaus. Finally, Nixon succeeded in getting the next attorney general, Robert Bork, to fire Cox. The result of this action was a massive and explosive reversal of public opinion resulting in demands for Nixon?s resignation or impeachment. (Muzzio 1982, 112-114) Most objective observers would proclaim that the Watergate cover up series of games was over, and it was time to resign, but Nixon believed that if he could keep the evidence (the tapes) from the public, he might still win. His choices at this point were to defy the court order and again propose the compromise of narrative summaries, appeal to the Supreme Court, or turn over the tapes. If the first option worked, he would be home free since he could make the summaries very general and ambiguous, which would validate the firing of Cox. The second would probably only delay impeachment and could result in a more restrictive constitutional interpretation for the office of the president. The third would end in impeachment. (Muzzio 1982, 117-121) Interestingly enough, Nixon?s response to defy the court order and propose the compromise would possibly land him both his best and worst result. Judge John Sirica was the one hearing the case. He had presided over the case from the beginning with the 113 Watergate burglars and was known as ?Maximum John? for often giving the maximum sentence possible for a crime. He was convinced that the burglars had not acted alone and was not sympathetic to Nixon. If Sirica did not cite Nixon in contempt, then Nixon could proceed with the narrative summaries. If Sirica did find him in contempt, then impeachment proceedings would immediately begin. (Muzzio 1982, 110-111) Sirica?s dominant strategy was to agree to accept the tapes (even though late) or to grant an extension to appeal (even though they were past the original deadline), but not to agree to any compromise. If Nixon understood Sirica?s dominant strategy, he would never have chosen to originally defy the court order. Eventually he turned over the tapes but with two inexplicably missing and another with an 18 1/2 minute gap. (Muzzio 1982, 112-114) According to Muzzio, as irrational as Nixon?s actions appeared at this point, he did act rationally based on what he believed were the dominant strategies of both Richardson (to stay) and Cox (to resign). He also misread the level of outcry from the public that, of course, would influence Congress. (Muzzio 1982, 112-114) At this point, with Cox gone and a compliant Attorney General now in place, Nixon?s dominant strategy changed from closing down the special prosecutor to avoiding impeachment. But this strategy resulted in Nixon?s worst outcome rather than his best because rather than having only the special prosecutor against him, he now had public opinion against him too. (Muzzio 1982, 112-114) The last game Muzzio discusses is between Nixon and the Supreme Court. In order to stall for time, Nixon appealed the subpoena of the tapes to the Supreme Court. He did not think he would win, but he thought that the three justices he nominated would 114 vote in his favor resulting in a divided court. If this happened, he could claim the partisanship of the court and refuse to hand over the tapes. He even hinted to the court that he might do this. If the court decision were unanimous, he had no choice but to hand over the tapes since he would be impeached either way. The Nixon appointees had to choose which way to vote. Muzzio thinks that they intentionally voted in unison against Nixon in order to have a unanimous decision that would discourage disobedience and avert a constitutional crisis. A better outcome for all would be for Nixon to comply, even with a divided court, but Nixon?s dominant strategy was to keep the tapes at almost any cost, and the court did not trust Nixon to comply with a divided court. (Muzzio 1982, 139-143) Muzzio determined Nixon?s dominant strategy based on Nixon's overriding motivation to obtain and retain power to the exclusion of all else. Also, he was a man who functioned in isolation by using a small staff to carry out his orders. These conclusions are based on his personal observations and authors such as David Abrahamsen, James Barber, William F. Buckley, Eli Chesen, Paul Halpern, Frank Mankiewicz, and Richard Neustadt, and by reading the personal accounts of those involved as well as official government reports. (Muzzio 1982, 156) As Muzzio points out, there were a lot of smaller games with individuals such as Cox, but the decision to cover up involved only Nixon and no other players. Opponents were Democrats, the FBI, and media, but their choices would be to investigate or not investigate. To investigate would be the worst case scenario for Nixon. If Nixon assumed that everyone would continue to investigate until the truth came out, then the 115 rational decision for Nixon would be to find out what really happened, blame over zealous employees of the CRP, and avoid an FBI investigation. Muzzio looked at various games throughout Watergate, but this study looks only at the presidential decision which was the point of no return from which Nixon could no longer recover--the decision which resulted in Nixon being forced to resign from the presidency. That was when Nixon learned about the illegal activities being planned and did not prevent them, or finding out after the fact about them and not allowing the facts of the case to be covered up. Much of what happened to get to the point of needing a cover up was the result of classic groupthink behavior (Janis 1967). Nixon established an atmosphere of pettiness where everyone was his enemy. The attitude that we have to get them before they get us was pervasive. When the police arrested the group at the DNC, many in Nixon?s inner circle initially felt that they had fallen victim to a Democratic or FBI conspiracy. (Magruder 1974, 197) Even the cover up on the part of Nixon?s people seemed to be the result of groupthink. Magruder sums it up when he says: ?No one even considered that there would not be a cover-up. It seemed inconceivable that with our political power we could not erase this mistake.? (Magruder 1974, 220) Despite this pervasive atmosphere that illegal actions equated to political hardball, Nixon still consciously made a decision that was part of a game where he chose an alternative that had the potential of a worst case scenario outcome. In Nixon?s game the other players were his co-conspirators. He had trusted them to make sure all evidence was hidden and to not tell the truth about what happened. This is a classic prison?s dilemma. 116 At the start of this particular game all players had picked the alternative of not cooperating with the authorities and staying silent. At that point Nixon thought that no one in the White House was involved but that some individuals in the CRP had known about and approved the break-in ahead of time. Based on this perception, Nixon had the following alternatives of the classic prisoner?s dilemma: Nixon?s Alternatives Keep silent Cooperate with authorities The other players had a similar set of alternatives: Other Player? Alternatives Keep silent Cooperate with authorities But since most of the players were not yet prisoners, they all had one more sub alternative under the ?keep silent? alternative. They could keep silent and do nothing, or they could keep silent and actively take action to cover up the crime, such as setting up one person to confess and take the blame, destroying evidence, and/or manipulating the justice system. A game matrix developed from these alternatives is shown in Table 6.1. 117 Table 6.1: Nixon Game Theory Matrix Other Players Stay Silent Cooperate Do Nothing Actively Cover Up Cooperate Do Nothing Nixon Stay Silent Actively Cover Up The rules of the game as Nixon saw them were that, in the series of games called Watergate, communication between players was allowed in an informal conspiratorial setting, but agreements between players were not necessarily binding. The players consisted of anyone who had direct knowledge of the break-in and cover-up. Players could make an unlimited number of moves any time they wanted to, making it a nonconsecutive game. It is unlikely that there would ever be a positive sum game, but the games would become negative sum if the players cooperated with authorities since they would still probably lose their jobs and receive some kind of jail sentence. The game would be zero sum if everyone remained silent. From his discussions with his staff and others, Table 6.2 describes how Nixon probably viewed the payoffs of different alternatives. 118 Table 6.2: Payoff of Different Options for Nixon over Watergate Cooperation Be Silent and Do Nothing Be Silent and actively Cover-up No possibility of arrest or impeachment for Watergate ......................... Other crimes or embarrassments Nixon was aware of could be discovered Possible negative impact on the upcoming election Little possibility of arrest or impeachment for Watergate Little chance of major damage before the election ........................ Other crimes or embarrassments will eventually be discovered Possible negative impact on Nixon?s second term power and legacy Little chance of major damage before election Other crimes or embarrassments may never be discovered Nixon?s power and legacy will remain strong ......................... Much greater possibility of arrest or impeachment Note: Benefits are listed above the dotted line under each column, and costs are listed below the dotted line. The alternative with the worst and the best payoffs would be to be silent and actively cover-up. Since Nixon perceived that other presidents managed to cover up similar incidents and get away with it, Nixon might have felt that he was just as politically astute as his predecessors. He not only wanted to win his second term election, he wanted to win in a landslide. He won his first term election by the skin of his teeth, and Nixon wanted to show his opponents that the voters loved and wanted him. But Nixon should have understood from the civil rights and the anti-war movements that the mood of the country was changing and that the voters were tired of being lied to by their leaders. It was ironic that Nixon campaigned on a law and order platform, and that the very type of judge he liked to appoint eventually would cause his downfall. Nixon should have also seen the futility of so many players keeping silent. A rational actor would always look for the least amount of effort for the best result. All players would prefer the other player to select payoffs in the following order: 119 be silent and actively cover-up, be silent and do nothing, cooperate. Based on these preferences, a prioritization of outcomes or results can be developed. For example, if Nixon were making a decision, purely as a rational actor, his best outcome would be for him to be silent and do nothing while all other players remain silent and actively cover up. This would be the least amount of effort for the best result and would therefore be Nixon?s number one priority. The same would be true for the other players. Basing the matrix values on this prioritization, each player, if they were rational actors, would probably prioritize as described in Table 6.3. Table 6.3: RAM Prioritization for Nixon-Other Players Nixon Priority Result (Nixon-Other Player) Other Player Priority 4 Silent and Do Nothing - Silent and Do Nothing 4 1 Silent and Do Nothing - Silent and Actively Cover-up 3 8 Silent and Do Nothing - Cooperate 5 3 Silent and Actively Cover-up - Silent and Do Nothing 1 2 Silent and Actively Cover-up - Silent and Actively Cover-up 2 9 Silent and Actively Cover-up - Cooperate 6 5 Cooperate - Silent and Do Nothing 8 6 Cooperate - Silent and Actively Cover-up 9 7 Cooperate - Cooperate 7 Based on the above prioritization the RAM matrix payoffs could therefore look like Table 6.4. 120 Table 6.4: Nixon-Other Player Game Theory Matrix under RAM Other Players Stay Silent Cooperate Do Nothing Actively Cover Up Cooperate 7,7 5,8 6,9 Do Nothing 8,5 4,4 1,3 Nixon Stay Silent Actively Cover Up 9,6 3,1 2,2 Under the RAM, a player should focus on the opponent?s action that would cause him or her the biggest problem. Therefore Nixon should have assumed that the other players would choose cooperate thereby giving Nixon his worst case scenario. Then, Nixon should have chosen the cooperate alternative to give him his own best worst case scenario. This clearly would have been the best alternative for Nixon since he still would have won the election, if not by such a landslide, and the legal actions would have probably remained at the level of the CRP. Nixon actually picked an alternative with his worst possible outcome. Although the other players initially chose to either be silent and do nothing or to be silent and actively cover up, they eventually changed their minds when faced with prison time and no help forthcoming from the White House. Judge Sirica, a law and order type judge, was ready to impose maximum sentences on each indicted conspirator in order to encourage cooperation. As players faced prison terms, their only rational alternative was to cooperate and negotiate as short a term as possible. This left Nixon with his worst case 121 scenario, demonstrating, that according to game theory, he had made an irrational decision. Determining a Different Utility Function Since Nixon?s Watergate decision did not follow the classic RAM, he must have either made his decision using a different utility function or he made a purely emotional decision that bore little resemblance to well considered rational thought. The first step in determining what his decision was based on is to look at what some authors of in-depth psychological studies believe were Nixon?s goals, objectives, and driving forces. Lammers and Genovese (2000) paint a picture of a man who wanted to be an activist president and who had an impact on both domestic and foreign affairs but primarily foreign affairs (Lammers and Genovese 2000, 224). He was actively involved in issues relating to welfare reform, crime, taxes, labor, abortion and civil rights, but many other domestic issues, such as the environment, were delegated to his cabinet (Lammers and Genovese 2000, 227). Nixon did not want his staff to be risk aversive. He was interested in taking positive, even risky, actions to get things done because otherwise he felt that nothing would be accomplished since he perceived that the Congress, media, and government bureaucracy were all against him (Lammers and Genovese 2000, 232). Pushing issues aggressively often worked to Nixon?s advantage because, in the long run, Nixon created more programs that lasted longer, such as the Environmental Protection Agency, than any president after him (Lammers and Genovese 2000, 245). Although Nixon could effectively work with the Russians, the Chinese and even leaders from the South, Nixon still viewed most groups as enemies or at the very least opponents. Whether one considers this trait as competitiveness or paranoia, at his core Nixon would 122 tend to divide people rather than unite them, (Lammers and Genovese 2000, 245) more along the lines of the Elitist Actor Model (EAM). Greenstein (2004) sees Nixon as a paradox. Nixon was both reclusive and political, cynical and constructive, politically astute and at the same time self destructive. He pushed a law and order platform while acting as if he was above the law. He trusted almost nobody, while at the same time he trusted over a dozen people to help him cover up several crimes (Greenstein 2004, 93). His keen political skill enabled him to make impressive gains, especially in the area of foreign policy and to package himself as the candidate who would represent the ?silent majority? while regularly belittling the same people he supposedly represented (Greenstein 2004, 107). Nixon never understood the people he represented, nor could he control his inner demons. Again, his style is closest to the EAM since we do not have a paranoid schizophrenic model to use (Greenstein 2004, 108). According the Barber (1992), Nixon?s key characteristic was his style of rhetorical manipulation. He won elections by carefully worded campaigns and debates backed up by good research on the preferences of his constituents and the shortcomings of his opponents. This style persisted throughout his career. Every challenge was dealt with by doing his homework, using heavy handed rhetoric, and making decisions without input or consensus by others (Barber 1992, 141). Although successful, rhetorical manipulation can not take the place of real concern, passion and conviction (Barber 1992, 165). According to Barber (1992), like Johnson, Nixon was an active-negative president who must continually try to keep his aggressiveness under control. This personality 123 comes off as being a fighter with whom voters often identify and a fighter is not supposed to give up easily. The problem is that there is a fine line between convincing someone to do something and applying force or manipulation. Presidents expect their orders to be followed without question. If a president who is the leader of the free world tells someone to take action, than de facto, the action must be the right thing to do. Nixon fell into this trap (Barber 1992, 485-486). According to Barber, the active-negative sees people as weak or grasping and as the primary threat. He consistently uses the same political style to manage all situations. Also, once a course is laid out the active-negative will follow it to the bitter end, not considering other alternatives (Barber 1992, 80-83). In Barber?s description, the EAM still stands out since Barber states that Nixon handles all situations that interest him as the sole decision maker, with little input, and no consensus. George and George (1998) agree with Barber?s assessment, in that Nixon developed a very rigid, formal style of relying only on himself to collect data, look at alternatives and make the best decision (George and George 1998, 213). Nixon could not develop real relationships due to his desire for privacy and his inability to accept criticism. The only people he could work with were those who provided order and loyalty. Nixon hated personal conflict and so surrounded himself with people who agreed with him and would not bring him bad news (George and George 1998, 214). His staff functioned as gate keepers, again evidence that Nixon falls closer to the EAM than any of the other models. Elitist Actor Model (EAM) According to Dye (2001), the EAM is a decision making model where large foundations established by corporate directors and other wealthy individuals fund think 124 tanks to identify problems and devise solutions that politicians then support. Money is the source of this elite power, and the motivation is to protect the source of the money and to retain power. If you control a large corporation, you have the money and ability to influence policy (Dye 2001, 41). Mills writes that the interlinking of three major institutions, the military, corporations and political parties, has placed enormous power in the decision makers of these institutions, and these decision makers have become the primary source of power in America (Mills, 1956). A person functioning under this model will take money from whatever source offers it and will take whatever action is necessary to maintain power. Unlike the Governmental Politics Model (GPM) decisions are not made through bargaining and compromise but primarily by the person in power or one of his designated gate keepers. Goals and objectives are consistent in that power must be protected at all costs (Dye 2001, 1-15). None of the illegal actions surrounding Watergate would ever have happened had the objective of being reelected with the highest margin humanly possible were not the one and only prime directive motivating all participants at the time. Although the players rationalized that they were following the political rules of the game, in that they took actions similar to those taken by other parties in other elections, they were actually making their own rules, as a king would do. The already established legal rules of the game were not rules to which they believed they needed to adhere. If Nixon?s primary object were to get reelected with the highest majority possible, giving him a strong mandate from the people that he could use to enhance his power, then he would be working within a much smaller objective timeframe than someone working under RAM. If reelection by a landslide were the prime objective, then loss of the 125 election would be the worst outcome for Nixon because he was not thinking past the election. Under an EAM, if Nixon lost his power, then nothing else mattered, but if he maintained his power, then he became invincible (like a king), especially if he won by a landslide. From an EAM viewpoint, the best decision would be to do whatever was necessary to retain as much power as possible -- to win the election by a landslide and then to deal with the fallout later. The problem with such an alternative is that Nixon was not king. He was president of a democracy and as the voters, media, and justice system learned the truth, they turned against him. As public opinion shifted, the other players, as a RAM would predict, started cooperating with authorities in order to protect their own interests. The EAM emphasizes sole decision making at the highest level without consensus and bargaining. Nixon should have understood, as he dealt with citizens expressing their disagreement with current government policies on the war and civil rights, that in a democracy, public servants can not act like elites even if they hold the highest office in the land. A democracy requires bargaining and consensus and following the rule of law. In Nixon?s mind, he was doing what many politicians had done before him, but what he did not realize was that he was no longer working in a gray area. He was not even straddling the line between tough politics and illegality. He had pole vaulted over this line and never even realized it. Using the EAM would only change Nixon?s priorities slightly. He still wanted the other players to keep silent and actively cover-up the crimes, but under EAM Nixon needed to actively make decisions. Reelection by a landslide was the primary goal and problems had to be contained until after the election. Assuming Nixon is functioning 126 under EAM and the other players under RAM produces the prioritization showed in Table 6.5. Nixon?s RAM priorities are included to facilitate comparison. Table 6.5: EAM Prioritization for Nixon-Other Players Nixon RAM Priority Nixon EAM Priority Result (Nixon-Other Player) Other Player RAM Priority 4 4 Silent and Do Nothing - Silent and Do Nothing 4 1 3 Silent and Do Nothing - Silent and Actively Cover-up 3 8 8 Silent and Do Nothing - Cooperate 5 3 2 Silent and Actively Cover-up - Silent and Do Nothing 1 2 1 Silent and Actively Cover-up - Silent and Actively Cover-up 2 9 9 Silent and Actively Cover-up - Cooperate 6 5 6 Cooperate - Silent and Do Nothing 8 6 5 Cooperate - Silent and Actively Cover-up 9 7 7 Cooperate - Cooperate 7 Based on the above prioritization the RAM matrix payoffs could therefore look like Table 6.6 below. Table 6.6: Nixon-Other Player Game Theory Matrix under EAM with all Alternatives Other Players RAM Stay Silent Cooperate Do Nothing Actively Cover Up Cooperate 7,7 5,8 6,9 Do Nothing 8,5 4,4 3,3 Nixon EAM Stay Silent Actively Cover Up 9,6 2,1 1,2 127 Even under the EAM, Nixon should have assumed that the other players would choose cooperate resulting in Nixon?s worst case scenario. Then, Nixon should have chosen the cooperate alternative to give him his best worst case scenario. This clearly would have been the best alternative for Nixon since he still would have won the election, if not by such a landslide, and the legal actions would have probably remained at the level of the CRP. Given the actions taken, it appears Nixon was operating in a much shorter duration worldview in order to reach his primary objective. Between June, when the break-in occurred, and November, when the election was held, it would have been very unlikely that anyone with knowledge of the cover-up would have cooperated with the authorities. If Nixon saw the objective and timeframe for this game as win the election by a landslide in November, then cooperate would no longer be an alternative, leaving the matrix to look like Table 6.7. Using this matrix, Nixon?s best choice is to actively cover up. Table 6.7: Nixon-Other Player Game Theory Matrix under EAM Other Players RAM Silent, Do Nothing Silent, Actively Cover Up Silent, Do Nothing 4, 4 3, 3 Nixon EAM Silent, Actively Cover Up 2, 1 1, 2 128 Conclusion In this case study, use of the EAM only explains Nixon?s decision when one understands that an EAM reduces the objective and timeframe to exclude the ?cooperate? alternative. So game theory can not fully be applied to a situation where the players? leadership style is based on an elitist actor model because all alternatives are not fully evaluated. Someone who views a situation, such as Watergate, from the viewpoint that the only goal is to gain another four years of power, will not considered any alternative that may result in the loss of power because an elitist assumes that with enough power, he or she will become invincible and above the law. Had the Watergate break-in come to light after Nixon?s second term election, Nixon would already be re-elected; his main objective would no longer be to win a second term but to remain in office. Under that scenario, cooperating with the FBI would probably have been considered a viable option by Nixon. Although EAM slightly affected the weightings of different alternatives in a game theory matrix, its primary effect was to totally eliminate a key alternative from consideration--the one with the worst case scenario--impeachment. 129 CHAPTER 7 WILLIAM JEFFERSON (BILL) CLINTON: MONICA LEWINSKY SCANDAL The three previous presidential mistakes have subtle ties to each other. Kennedy?s Bay of Pigs mistake drove him deeper into a cold war with the USSR. It was Kennedy who allowed 15,000 military advisors to be stationed in Vietnam and was influential in the overthrow of the South Vietnam leader Ngo Dink Diem (Schlesinger 1965, 992). Johnson not only continued but enhanced Kennedy?s military presence. Some of Nixon?s Watergate mistakes can be tied to his close loss to Kennedy in the 1960 election. Nixon was so determined to win his third election in a landslide that he made numerous irrational mistakes. William Jefferson Clinton?s mistake stands apart from the other big presidential mistakes in several ways. Other than its impact on the office of the president and the ability of the Democrats to press their agenda, Clinton?s initial lapse in judgment did not harm anyone but Clinton, his family, and the women involved. Unlike Kennedy?s and Johnson?s mistake, nobody was physically harmed, and vast sums of government money were not spent. Unlike Nixon?s mistakes, numerous people did not end up in jail. Clinton?s mistake was not the result of great historical events overwhelming a president not quite up to the task. It was not the result of faulty advisors doing things they should not have done or providing bad information, or not speaking up when they should have. Clinton?s only advisor was himself. Clinton?s mistake had nothing to do with his staff, 130 the voters, national interests, or historical events. Finally, Clinton?s cheating on his wife was not illegal and was not an act that had not been committed by numerous other presidents. Clinton?s mistake was trying to cover-up this lapse in judgment. It was similar to Nixon?s cover-up mistake except that Clinton had committed the act he was trying to cover-up and Nixon had not. Yet in Nixon?s case the act was a crime and in Clinton?s it was not. One wonders how someone as smart and politically savvy as Clinton could have gotten impeached for trying to cover-up an act that, in light of normal presidential actions, was insignificant? Historical Context Clinton was the 42nd President of the United States. He served two terms from 1993 to 2001 and left office with the highest approval rating (65%) of any president since World War II. Clinton was a Democrat from Arkansas where he served two terms as governor prior to being elected president. He was the third youngest person to ever be president after Theodore Roosevelt and John F. Kennedy. He was the first president to represent the baby boomer generation that was born right after World War II. During his two terms the U.S. transitioned from a cold war mentality into the longest period of peace and economic expansion in history, resulting in a balanced budget and surplus funds. Clinton was not always able to press his agenda, especially when the Republicans won control of the House of Representatives in 1995 as a result of Clinton administration policy failures such as health care reform. Clinton?s second term was engulfed by scandal and impeachment. (Greenstein 2000, 173-190; Lammers and Genovese 2000, 299-328) 131 Clinton and Lewinsky In July 1995 the White House Chief of Staff?s office hired a new intern named Monica S. Lewinsky (Clinton 2005, 407). Lewinsky was a bright, attractive, 22 year old college graduate, from a prosperous Jewish family in California. During her internship, she met Clinton at a White House office party in November 1995. While flirting with him during the party, Lewinsky showed Clinton the top of her thong underwear and Clinton invited her to his private study. That evening and again two days later, Lewinsky performed oral sex on Clinton while Clinton talked to members of Congress on the telephone (Baker 2000, 81; Isikoff 2000, 171; Posner 1999, 17; Thompson 2000, 151). A few days later Lewinsky moved to a paying job with the White House Office of Legislative Affairs. This new position had been offered and accepted weeks prior and had nothing to do with Clinton (Posner 1999, 16). It was not until New Years? Eve that they had another sexual encounter after Lewinsky approached Clinton in his private dining room at which time she was obliged to remind him of her name. Then in early 1996 Clinton approached Lewinsky three times over a one month period for intimate relations (Clinton 2005, 408; Posner 1999, 18). During this period, and until the end of their friendship, Clinton and Lewinsky exchanged numerous gifts (Baker 2000, 81; Isikoff 2000, 171; Clinton 2005, 408). On February 19th Clinton told Lewinsky that she did not want to continue the affair, but a little over a month later called her for another sexual meeting. In April Lewinsky?s supervisors in the White House Office of Legislative Affairs were concerned that she was spending too much time around the President and not enough time doing her 132 job. As a result they transferred her to the Pentagon as secretary to the public affairs officer. This new job was of slightly higher pay, but Lewinsky wanted to remain in the White House (Clinton 2005, 407; Isikoff 2000, 210; Posner 1999, 18). During their next sexual encounter, Lewinsky complained of this transfer to Clinton, and he told her he would transfer her back to the White House after he won the 1996 election. With Lewinsky in the Pentagon and Clinton running for his second term, they did not see each other again for over a year but occasionally talked on the phone late at night (Baker 2000, 82; Clinton 2005, 408; Klein 2002, 5). Then in February 1997, after the taping of Clinton?s weekly radio speech Clinton had his secretary Betty Currie ask Lewinsky not to leave and Clinton and Lewinsky had sexual relations in a study off the Oval Office (Clinton 2005, 408). This was the meeting where Clinton got his semen on Lewinsky?s blue dress that she later hung in her closet unlaundered (Baker 2000, 81). Two more times Clinton asked Currie to arrange meetings between Clinton and Lewinsky; the first in March for a sexual encounter and the second one in May when Clinton finally ended the affair for good, although they continued to exchange calls and gifts. Clinton claimed to have ended the affair out of guilt about his family but offered to be her friend and provide assistance (Clinton 2005, 409; Isikoff 2000, 171; Posner 1999, 19). Two days after the second break up, the Supreme Court ruled that Paula Jones? civil suit against Clinton could proceed even though he was a sitting president. Paula Jones was an Arkansas state employee who was suing Clinton for sexual harassment because Clinton had propositioned her and exposed himself to her in a hotel room. 133 Clinton had tried to delay the suit saying that he should not have to deal with civil suits until he was out of office (Baker 2000, 23; Isikoff 2000, 103; Thompson 2000, 151). In spite of their break up, Lewinsky insisted that Clinton keep his promise to transfer her back to the White House and contacted him several times during the next few months about that subject (Clinton 2005, 409; Posner 1999, 19). In the meantime, she had confided about their relationship to her co-worker Linda Tripp, a 48 year old political appointee from the Bush administration who worked in the same office in the Pentagon as Lewinsky. Tripp did not like the Clinton administration and had discussed writing a tell-all book about Clinton with a conservative literary agent. This same literary agent advised Tripp to record her conversations with Lewinsky to protect herself in case she had to testify. Tripp also convinced Lewinsky not to wash Clinton?s semen off her dress (Baker 2000, 137; Clinton 2005, 409; Isikoff 2000, 175; Posner 1999, 22; Thompson 2000, 152). In October 1997, Jones? lawyers received an anonymous tip, almost certainly from Tripp or her agent, about Clinton?s affair with Lewinsky. By December Lewinsky was called to make a deposition as a witness for Paula Jones (Posner 1999, 20). Tripp was already on the witness list because she had seen Kathleen Willey, another woman claiming sexual harassment by Clinton, depart from Clinton?s office. Tripp had told a reporter that Willey appeared disheveled and happy (Isikoff 2000, 175). Although Clinton promised to bring Lewinsky back to the White House if he was re-elected, he made no effort to do so. When Lewinsky indicated that she might like a job in New York, Clinton jumped at the chance to get her out of Washington, D.C. and secured her a job offer at the United Nations. When she refused the United Nations job, 134 Clinton asked his friend Vernon Jordon to look for an even better position for her. In January 1998 Jordon found her a job with Revlon and Lewinsky accepted (Baker 2000, 81; Clinton 2005, 409; Posner 1999, 22). In December 1997, when Clinton found out that Lewinsky would testify in the Jones case, he advised her to justify her numerous visits to his office by saying that she was meeting his secretary Currie. He also recommended to her that if she submitted an affidavit saying she was not sexually harassed by Clinton, she might avoid any further involvement in the Jones case (Baker 2000, 81; Clinton 2005, 409; Posner 1999, 20). Clinton never directly told Lewinsky to lie. Neither did Clinton ask Currie to lie although he provided her with a false series of events, making statements along the lines that he was never alone with Lewinsky and that Lewinsky was the aggressor and not him (Baker 2000, 91; Posner 1999, 25; Thompson 2000, 153). Clinton asked Currie to collect all the gifts Clinton gave to Lewinsky. She collected them and hid them in her home (Baker 2000, 81; Isikoff 2000, 261; Posner 1999, 21). In January 1998, Jordan coached Lewinsky on her testimony and prepared an affidavit for her denying any sexual encounter with Clinton. It was signed and sent to Jones? lawyer (Isikoff 2000, 265; Posner 1999, 21; Thompson 2000, 153). When Clinton was deposed for the Jones trial, he denied any sexual relations with Lewinsky. Using the strict definition of sexual relations provided by Jones? lawyers, Clinton was able to partially tell the truth since Monaca had performed oral sex on him and the definition required an action on his part (Clinton 2005, 407), but since Clinton had also fondled Lewinsky, he was not only making misleading statements but was also lying (Posner 1999, 25). He also lied when he stated that he could not recall ever being alone with 135 Lewinsky or ever giving her gifts except maybe a hatpin (Baker 2000, 80; Posner 1999, 25; Thompson 2000, 153). Also, since he had ended their relationship, he could honestly say that they were not in any kind of sexual relationship. Although technically true, and arguably not perjury, it was certainly misleading (Isikoff 2000, 369; Klein 2002, 163; Posner 1999, 22). Tripp gave Independent Counsel Kenneth Starr the recordings she made of Lewinsky?s phone calls talking about her relationship with Clinton. Starr had been appointed independent counsel in 1994 to investigate whether the Clintons? involvement in a failed real estate venture called Whitewater was fraudulent. Starr?s investigation later expanded to include other Clinton activities (Clinton 2005, 246). Lewinsky, unaware that Tripp had met with Starr, tried to convince her friend to lie at the Jones trial by giving her a memorandum about what Tripp should say at the deposition (Baker 2000, 167; Posner 1999, 24; Thompson 2000, 153). Caught red handed on Tripp?s recorder and facing charges of perjury and witness tampering, Lewinsky accepted an immunity deal from Starr in July 1998 in exchange for her full cooperation, which included producing the unwashed blue dress with Clinton?s semen on it (Clinton 2005, 439; Posner 1999, 28). DNA testing confirmed that the semen was Clinton?s. In August 1998 Clinton had no alternative but to admit that he had misled the judge for Jones? trial, had misled the American people, and had lied to his family and friends (Isikoff 2000, 221; Baker 2000, 23; Stucky and Shannon 2000, 519; Thompson 2000, 155). On December 19, 1998, the Republican led House of Representatives impeached Clinton on two articles of impeachment (perjury and obstruction of justice) making 136 Clinton the first elected president to ever be impeached. The Senate trial lasted a little over a month. The Republicans could not even get a majority to vote against Clinton, let alone the 2/3 majority needed for removal. On February 12, 1998 the Senate acquitted Clinton on both counts, thus saving Clinton?s presidency but leaving him a disastrous legacy which had not been seen since Nixon?s Watergate legacy (Baker 2000, 436; Thompson 2000, 155). Rational Actor Model This was a classic prisoner?s dilemma except for one thing. Clinton was unaware that Lewinsky had confided their relationship to Tripp, especially since she had told him that she had not. Clinton thought that he was dealing with only two other players who knew the truth?Currie and Lewinsky, both of whom he thought he could trust. Currie was a loyal worker, and Lewinsky was in love with him, plus he had secured Lewinsky an enviable job with Revlon. Perhaps Clinton should have been suspicious when Lewinsky was called as a witness in the Jones suit. Although Clinton did tell Starr that he knew Lewinsky would eventually tell someone (Klein 2002, 206), he probably did not expect it to happen so soon or so publicly or with such definitive evidence as semen on a dress. Clinton purposely never had intercourse with Lewinsky as a safeguard against discovery and did not ejaculate until the last two of the ten sexual meetings (Posner 1999, 17). It was to everyone?s advantage to remain silent except for the unknown fourth player, Linda Tripp. Tripp had absolutely no motivation to keep quiet. She was not particularly attached to Lewinsky, having known her only a year and a half, and she saw great personal advantage in telling the world her story through books and interviews. 137 During Clinton?s campaign for his first term as president, he skirted rumors of his long time affair with Gennifer Flowers by neither confirming or denying it but by stating that he was sorry to have had caused his family pain. This strategy worked well because it showed remorse while letting the media and the public know that this was a private matter between him and his family (Thompson 2000, 150). Had Clinton not been forced to testify in the Jones law suit, he may have handled the Lewinsky affair the same way, thereby eliminating the possibility of a worse case scenario of being removed from office. During the impeachment proceedings, Clinton was charged with perjury based on his statements in the Jones deposition where Clinton was not fully truthful under oath when asked about his relationship with Lewinsky. Most of Clinton?s obstruction of justice charge also had to do with the Jones deposition when he asked Lewinsky to return his gifts and provided Currie with a false series of events concerning his assignations with Lewinsky. Finding Lewinsky a job was probably done more to keep her from pressuring him to bring her back to the White House than an effort to keep her from telling the truth. The Jones deposition game was not a consecutive move game where Clinton made a move and the other players responded or where Clinton would respond to the moves of the other players. In this game Lewinsky made moves in response to what she thought the president would do (lie about the relationship), and Tripp made moves in response to what Lewinsky was doing (lying and trying to get Tripp to lie also). Currie?s moves were based on what Clinton asked her to do as long as they were legal. In Clinton?s mind, none of the players had any real evidence, and the two players who knew the facts would not betray him. Clinton knew Lewinsky had signed an 138 affidavit saying she had not had a sexual relationship with him which prevented her from changing her story without admitting to perjury. Currie knew about the gifts and about Clinton being alone with Lewinsky but could not testify to a sexual relationship since she had not observed one. So, Clinton believed that he was covered. In this game, the only alternatives Clinton could make were the following: Clinton Alternatives: Lie Tell the truth When the Supreme Court ruled that the law suit could go forward, Clinton had to testify. He could not refuse to speak as he did with the media concerning Gennifer Flowers because he was being asked questions in a courtroom setting and taking the fifth amendment would not be appropriate since having sex with Lewinsky was not a crime. So, he could not keep quiet for the purpose of not incriminating himself. The other players had similar alternatives: Alternatives of Other Players: Lie Tell the truth A game matrix developed from these alternatives is shown in Table 7.1. Table 7.1: Jones Deposition Game Theory Matrix Other Players Lie Tell the Truth Lie Clinton Tell the Truth 139 The rules of the game as Clinton saw them were that the Jones deposition game was a one time nonconsecutive game. Communication between players flowed both ways, and agreements between players were possible but nonbinding. Clinton thought that players could make a move only when they testified. This was primarily true for Clinton, Lewinsky, and Currie, but Tripp did not play the game according to the rules Clinton was expecting. In fact, Clinton did not even see Tripp as a player. Tripp taped her conversations with Lewinsky and then went directly to Starr with the evidence. Table 7.2 below describes how Clinton probably viewed the payoffs of different alternatives. Benefits are listed above the dotted line under each column; costs are listed below the dotted line. Table 7.2: Payoff of Different Options for Clinton Concerning the Jones Deposition Lie Tell the Truth Less chance of losing law suit Not embarrassed before public and family ......................... Could be prosecuted for perjury leading to possible impeachment, removal from office, and jail Did nothing illegal to jeopardize job and freedom ........................ Greater chance of losing law suit Embarrassed in front of public and family Could cause Lewinsky to be charged with perjury One might argue that telling the truth resulting in a possible divorce from his wife might be a worse case scenario that the loss of the presidency, but Clinton had cheated in the past and the marriage had survived so it is unlikely that this short dalliance with Lewinsky would have caused a breakup. 140 Table 7.3 below describes how the other players probably viewed the payoffs of different alternatives. Benefits are listed above the dotted line under each column; costs are listed below the dotted line. Table 7.3: Payoff of Different Options for the Other Players Concerning the Jones Deposition Lewinsky Currie Tripp Lie Tell the Truth Lie Tell the Truth Lie Tell the truth Would not jeopardize her job Would not jeopardize Clinton friendship ................ Tripp might repeat her confidences Prosecuted for perjury Would not commit perjury Would not matter what Tripp did ................ Could jeopardize her job Could jeopardize Clinton friendship Would not jeopardize her job unless discovered Would not jeopardize her loyalty to Clinton ................ Prosecuted for perjury Would not commit perjury ................ Might jeopardize her job Would not demonstrate loyalty to Clinton People would believe her if others lied Support her friend Lewinsky ................ Clinton would not be punished Would hurt Clinton ................ No one would believe her if others lied and Lewinsky destroyed the blue dress Would hurt her friend Lewinsky For Clinton, the alternative with the worst payoffs would be to lie. He had already won his second term; the loss of the presidency would not be at risk with the truth but it would be with a lie. Also, the truth could lose him the law suit and would embarrass him, but a lie could lose him the presidency and his law license and could even mean jail time. Since he was unlikely to lose his family by telling the truth, lying was the alternative with the worst payoff. 141 A rational actor would always hope for the least amount of effort for the best result. Clinton would prefer the other players to select payoffs in the following order: lie, tell the truth. Lewinsky would prefer Clinton to lie since she had already done so with her affidavit and did not want to be charged with perjury. The other two players would hope for Clinton to tell the truth, so that they could also tell the truth and be believed. Based on these preferences, a prioritization of outcomes or results can be developed. For example, if Clinton were making a decision purely as a rational actor, his best outcome would be for him to lie and for all the players to lie also. Typical of a prisoners? dilemma, this would be the least amount of effort for the best result and would therefore be Clinton?s number one priority. For Tripp the lowest priority would be for Clinton to lie so she would not be believed when she told the truth unless she had incontrovertible evidence against him. Then, her lowest priority would be for Clinton to tell the truth and for her to lie. Basing the matrix values on this RAM prioritization, each player, if they were rational actors, would probably prioritize like Table 7.4 below. Table 7.4: RAM Prioritization for Jones Deposition Game Clinton Priority Result Lewinsky Priority Currie Priority Tripp Priority 1 Lie-Lie 1 3 3 2 Tell truth-Tell Truth 2 1 2 3 Tell truth-Lie 4 4 4 4 Lie-Tell truth 3 2 1 Based on the above prioritization the RAM matrix payoffs could therefore look like Table 7.5 below. 142 Table 7.5: Jones Deposition Game Theory Matrix Lewinsky Currie Tripp Lewinsky Currie Tripp Lie Tell Truth Lie 1,1 1,3 1,3 4,3 4,2 4,1 Clinton Tell Truth 3,4 3,4 3,4 2,2 2,1 2,2 Under the RAM, Clinton should assume that the other players would choose to tell the truth to give Clinton his worst case scenario. Then, Clinton should choose the tell the truth alternative to give him his best worst case scenario. One problem is that Clinton did not know Tripp was a player, but subtracting her from the matrix above does not change the result under RAM. Lewinsky?s first priority was to lie, and Currie had only minimally damaging knowledge. So Clinton probably thought lying was an acceptable risk even though it did not follow the RAM. In Clinton?s mind, there were only two players, and both were motivated to not incriminate him. Determining a Different Utility Function Since Clinton?s Jones deposition decision did not follow the classic RAM, he must have either made his decision using a different utility function, or he made a purely emotional decision that bore little resemblance to well-considered rational thought. The first step in determining what his decision was based on is to look at what some authors of in-depth psychological studies believe were Clinton?s goals, objectives, and driving forces. Lammers and Genovese (2000) paint a picture of a man who has a high level of political skill. Clinton took great notice of polls and focus groups to get a feel for public reaction and worked with members of Congress to understand what Congress would 143 tolerate (Lammers and Genovese 2000, 310). He was quick to modify his policies in order to achieve at least some measure of his policy goals. It was surprising how much Clinton was able to accomplish despite a divided government (Democratic presidency and Republican Congress). Clinton was intelligent, energetic, ambitious, self-confident, and charismatic. He showed great skill in running his reelection and in passing legislation (Lammers and Genovese 2000, 303). From this analysis the governmental politics model (GPM) seems to be the most descriptive of the Clinton personality. Greenstein (2004) sees Clinton as a totally political animal in both his public and private life. Greenstein also describes Clinton as smart, articulate, energetic, undisciplined, and effective under pressure (Greenstein 2004, 174). He could fluidly adjust his rhetoric to the political context and easily recover from small political missteps. His self confidence and political skill, along with the booming economy, kept him popular with the public through most of his administration (Greenstein 2004, 174-178). George and George (1998) describe Clinton?s management style as a blend of ?collegial? and ?formalism.? Clinton supported broad participation and open communications among his cabinet and staff. His organization also had the effect of limiting his engagement to only those issues he perceived were most important (George and George 1998, 241-244). Clinton wanted to be informed of what was going on, but he hired advisors who could take action without his constant supervision (George and George 1998, 249). George and George describe Clinton as very intelligent, open minded, energetic, cool under pressure, and a person who looked carefully into all aspects of an issue before making a decision. Clinton loved political conflict and would disarm opponents by pointing out positive aspects of both sides of the debate before announcing 144 his preferred policy. He was not as good at personal conflict and tried to avoid confrontation (George and George 1998, 247) but was eager to tackle political issues that others might avoid such as welfare and health care reform (George and George 1998, 248). George and George also describe Clinton as a political animal, always ready to compromise in order to find a consensus that would make his policy viable (George and George 1998, 248). All the authors are in agreement that above all else, Clinton?s personality and actions are highly political in nature. Therefore, there is consensus that Clinton follows the governmental politics model (GPM) and does not seem to come even close to any other model. Governmental Politics Model (GPM) In the GPM, decisions are made through bargaining and compromise. Goals and objectives are not consistent but depend on the strength of specific interest groups, public opinion, and constituency ideology. The rules of the game are the political rules first and foremost, which means that fund raising and keeping constituents happy are keys to staying in power (Allison and Zelihow 1999, 255-313). If Clinton?s primary objective was to please constituents, interest groups, and party members in order to stay in power, he would also be interested in not angering the public or his family. Clinton had a pattern of sexual indiscretions, and when caught would hide the details from the public and the media. This worked quite well in the past. From a GPM viewpoint, a pattern of hiding embarrassing personal activities had worked the best for Clinton. We are assuming here that the activities themselves were something that Clinton could not or would not control. Telling the public as little as possible 145 continued with the Jones deposition game even though, as a lawyer, he should have been aware of what could happen if he was caught lying. Clinton?s best decision would be to tell the truth, but as a lawyer he also knew, or thought he knew, that what he did with Lewinsky could not be proven. Even if she told the truth, he could always portray her as an obsessed stalker who had to be transferred to the Pentagon for her own good. In Clinton?s memoirs he never admits to perjury. He admits to making misleading statements that were technically true. In the world of politics, a spin can be put on almost any statement to make it technically true, but Clinton said that he did not remember being alone with Lewinsky and that he only remembered giving her one gift and that he did not have sexual relations with her. In a political context these statements could be considered somewhat true, but in the cold hard light of the justice system, Clinton committed perjury. Clinton knew he lied because it is unlikely that he would have denied his actions with Lewinsky had he known that she had never washed the semen-stained dress. The GPM emphasizes bargaining and compromise. By Clinton?s standards of political behavior, he testified appropriately because he thought his answers could not be disproved. Clinton said that he did not remember being alone with Lewinsky and he thought he may have given her one gift. Clinton thought that only Lewinsky could know the full truth and she had already sworn that they had not had sexual relations plus she had already accepted a job that Clinton was largely responsible for getting for her. Even if she admitted what really happened it would be her word against his. Under GPM the matrix provided below does not change, it still looks like the matrix shown in Table 7.5. 146 Table 7.5: Jones Deposition Game Theory Matrix under Clinton Utility Function Lewinsky Currie Tripp Lewinsky Currie Tripp Lie Tell Truth Lie 1,1 1,3 1,3 4,3 4,2 4,1 Clinton Tell Truth 3,4 3,4 3,4 2,2 2,1 2,2 The matrix does not change because Clinton lacked key information about the players. Clinton did not think he needed to tell the truth because as a lawyer he did not see where there was any significant evidence against him. Clinton had not practiced law for a long time, but he knew that his guilt would be hard to prove and admitting guilt would damage him politically and personally and possibly cause him to lose the Jones case. Clinton did not follow the RAM model because, due to a lack of information and his political motivations, he thought he would get away with lying and avoid the negative consequences of his actions with Lewinsky. Conclusion In this case study, use of the RAM does not explain Clinton?s decision to commit perjury during the Jones deposition but a modified utility function helps to explain the decision by reorienting our perception of what Clinton did based on a governmental politics model (GPM) and a lack of key information. Clinton gave his testimony in a way that he thought could not be disproved. Clinton did not follow the RAM because he believed that there was little chance of being fount out. If he had known about Linda Tripp or the unwashed semen-stained dress, Clinton would probably have followed the RAM and told the truth. 147 CHAPTER 8 ANALYSIS, SUMMARY AND CONCLUSION The central hypothesis of this study was that decision making models can be integrated into the rational actor model (RAM) where the RAM would still retain its integrity while providing a better explanation of individual political actions so when using these other decision making models with the RAM, a more complete explanation of decisions can be provided than by using the RAM alone. To test this hypothesis, a qualitative research design was used. This type of approach is necessary for game theory since many details must be collected on each game in order to predict how the game will be played using the RAM. These details are also needed to objectively select an appropriate model for the determination of a utility function. Allison and Zelihow?s (1999) two decision making models, the governmental politics model (GPM) and organizational behavior model (OBM), plus Dye?s (2001) elitist actor model (EAM), were used to determine if the results can be better explained with the help of these models. The challenge in applying game theory is determining how a player reads other players and how he or she evaluates payoffs. These other models were used to shed light on these issues. Using the McConnell Center?s list of top ten mistakes and this study?s criteria of only cases since (and including) John Kennedy, five cases result. The cases are listed in Table 8.1. 148 Table 8.1: Presidential Mistakes Making the McConnell Center?s Top Ten list and Occurring between 1961 and 2000 President?s Name Mistake Survey Ranking John Kennedy Bay of Pigs Invasion 8 Lyndon Johnson Escalation of the Vietnam War 3 Richard Nixon Watergate 5 Ronald Reagan Iran-Contra Affair 9 Bill Clinton Monica Lewinsky Scandal 10 (Source: Gregg, 2006) All of these mistakes were used as cases in this study except the Iran-Contra Affair since it has never been proved that Reagan knew about the situation, let alone made a conscious decision about it (Walsh 1993, xiii). Each case study contained a description of the situation, including historical context and the actions of each of the major actors. Then, the event was analyzed as a game using classical RAM theory where the alternatives and objectives of all players were listed. Four well-respected books on presidential psychological analysis were used in order to determine which model to use to inform the game theory analysis. These books were by Barber (1985), George and George (1998), Lammers and Genovese (2000), and Greenstein (2004). From these four books a consensus on each president's decision making style was determined by summarizing what each book says is the ex- president?s style and comparing these summaries to the three different decision-making models. The model used was the model that had the most traits in common with each 149 book?s summary. After each decision was analyzed using game theory, it was again analyzed using game theory informed by the appropriate decision-making model based on the ex-president's agreed-upon decision-making style. The purpose was to determine if the alternative model chosen could better explain the presidential ?mistake.? Table 8.2 outlines each of Allison and Zelihow?s models, the EAM, and the RAM. Table 8.2: Key Elements of the Decision-making Models Applied to the Game Theory Process Rational Actor Model (RAM) Organizational Behavior Model (OBM) Governmental Politics Model (GPM) Elitist Actor Model (EAM) Goals/Objective ? Gain advantage over other players ? Win ? Grow the organization ? Stay in power ? Raise money ? Please constituents ? Win elections ? Retain power ? Grow power Alternatives ? Anything is possible ? Decision made by a single player in a deliberate manner ? Limited by o Organizational capabilities o Standard operating procedures o Organizational Culture ? Limited by alternatives that o Would win elections o Leave a legacy ? Limited to relying on other elites Outcomes/payoffs ? Straight forward ? Based on specific unit of measure or a prioritizing of alternatives ? Accomplish organizational mission ? Win elections ? Maintain position of leadership ? Move up in leadership positions ? Stay in power ? Retain power ? Grow power Strategy ? Minimize risk and maximize advantage ? Noncooperative with other organizations ? Small choices made by multiple players ? Cooperative ? Bargaining ? Noncooperative with other actors ? Secretive from other actors Rules of the game ? Depends on the game ? Keep the boss happy ? Take actions to get promoted ? Follow standard operating procedures ? Be a team player ? Loyalty to organization ? Legal rules ? Formal rules ? Informal rules, e.g., log rolling ? Cooperative only with other elites Based on Allison and Zelihow (1999) and Dye (2001) and Mills (1956) 150 The material that follows presents a brief review and summary of each of the four cases discussed in this study. Kennedy and the Bay of Pigs When Kennedy made the decision to proceed with the Bay of Pigs operation to establish a beachhead on Cuba in an effort to overthrow Castro, he seemed to be making a decision totally at odds with the RAM. The Cuban people were the other player and the success of the Bay of Pigs operation depended on the uprising of the Cuban people against Castro. The operation would not work without Cuban support because so few people were sent, and they had no U.S. military backup. Rather than assuming a worst case scenario of the Cuban people not rising up against Castro, Kennedy seemed to assume a best case scenario when he made his decision, which was counter to the theory behind the RAM. Kennedy did not match up with any of the other models, but he seemed to be a man who had a strong motivation to win and to do what seemed right for others (U.S. national security and the Cuban exiles). The one alternative that would have worked well with these motivations under the RAM would have been to redirect the exiles from an attack to an infiltration operation. Unfortunately the CIA provided Kennedy with poor information by telling him that if the exiles could not maintain a beachhead, they could disappear into the mountains and take on the infiltration mission instead. So in Kennedy?s mind, executing the operation included both attack and infiltration even though this was not realistically possible since the mountains were too far away and on the other side of impassible swamp land. 151 Kennedy could not make a rational decision without accurate information, yet, to some extent, at least by his own perception of his choice, he did make a decision that fit in with RAM using his own utility function of winning and doing what he thought was right. Of all the presidential mistakes discussed, Kennedy made his decision base on the most moral of intentions, yet it still won him the honor of later being judged to have made the eighth worst U.S. presidential mistake in history. Johnson and the Escalation of the Vietnam War Johnson did not simply decide to escalate the Vietnam War without a more comprehensive plan in mind. He actually picked two alternatives: escalation (to demonstrate strength) and then diplomacy. Johnson was trying to send a message to Ho Chi Minh, the leader of North Vietnam, that he must agree to negotiate or face the wrath of the U.S. military. Johnson was trying to force Minh to let South Vietnam stay independent, but Ho Chi Minh?s motivation was to free his country from all colonizers. To Minh, the U.S. was no different from the Chinese or the French when they tried to control Vietnam in years past. Johnson followed the GPM and projected his political motivations onto a man whose primary motivations were nationalistic. When Minh would not negotiate, Johnson had no alternative left except to escalate because once the U.S. Air Force bombing had begun, Johnson?s self esteem was on the line. If the North would not agree to his offer of an economic package in exchange for leaving the South in peace, he had to back up his threat of military force or look weak and naiv? in the eyes of the international community and the American public. Although Johnson seemed to make a decision that was not rational because it had the possibility of the worst harm to the U.S., it was actually 152 rational because Johnson was not necessarily seeking to avoid the worst outcome for the U.S.; he was seeking to avoid the worst outcome for Lyndon Johnson. The most embarrassing immediate outcome was orchestrating the first U.S. defeat and making the U.S. look weak and impotent, and by extension making Johnson look weak and impotent. The North not only would not negotiate, but they escalated the violence directly against the U.S. military. If the North had only acted against the South, then Johnson could have continued to provide limited military support, but the North kept targeting U.S. personnel and since Johnson already had congressional support to escalate, he had painted himself into a corner once the North rejected his overture of diplomacy. Every U.S. life that was lost brought more and more pressure on Johnson to win the war. Johnson did not want the country or his party, and by extension himself, to appear weak in the face of an enemy, especially one with a small third world economy. He made it clear that he would not be the first president to lose a war (Logevall 1999, 392). Johnson made decisions that committed the U.S. to a long and lengthy war if Minh refused to negotiate. When Minh would not give him the one thing he had to have to win, a non-Communist South Vietnam, Johnson could not back away. He had not left himself any alternatives except to slowly drag the U.S. into a protracted war and win the honor of being the president who made the third worst mistake in the history of this country. Nixon and Watergate Nixon?s motivations followed the EAM. He faced a classic prisoner?s dilemma where he was depending on the silence of numerous people to keep the public and the 153 special prosecutor from finding out about his cover-up. To depend on the indefinite silence of so many others was foolhardy under the RAM and proved to be so in actuality as events unfolded. As an elitist, Nixon?s primary objective was to get reelected with the highest majority possible. A landslide election would give him a strong mandate from the people that he could use to enhance his power. Because of this Nixon was working within a much smaller objective timeframe than someone working under RAM. If reelection by a landslide were the prime objective, then loss of the election would be the worst outcome for Nixon because he was not thinking past the election. Under an EAM, if Nixon lost his power, then nothing else mattered, but if he maintained his power, then he became invincible (like a king), especially if he won by a landslide. So Nixon?s best decision was to do whatever was necessary to retain as much power as possible -- to win the election by a landslide and then to deal with the fallout later. Unfortunately for Nixon he was not king. He was president of a democracy and as the voters, media, and justice system learned the truth, they turned against him. As public opinion shifted, the other players, as the RAM would predict, started cooperating with authorities in order to protect their own interests. Given the actions taken, it appears Nixon was operating in a much shorter time frame in order to reach his primary objective. Between June, when the Watergate break- in occurred, and November, when the election was held, it would have been very unlikely that anyone with knowledge of the cover-up would have cooperated with the authorities. Since Nixon?s key objective was winning the election by a landslide, cooperation was no 154 longer an alternative, sealing Nixon?s fate as making the fifth worst presidential mistake in U.S. history. Clinton and the Lewinsky Affair Clinton, like Nixon, also faced a classic prisoner?s dilemma but with far fewer players and for a cover-up of actions which were not even illegal. Perhaps Clinton believed lying was an acceptable risk even though it did not follow the RAM. In Clinton?s mind, there were only two players, and both were motivated to not incriminate him. Like Johnson, Clinton followed the GPM. In the past, when caught in sexual indiscretions he would hide the details from the public and the media. Telling the public as little as possible continued with the Jones deposition even though, as a lawyer, he should have been aware of what could happen if he was caught lying. Clinton?s best decision would be to tell the truth, but as a lawyer he also knew, or thought he knew, that what he did with Lewinsky could not be proven. Under a GPM Clinton thought he could lie at the deposition and never be caught since he had reached a kind of bargain with Lewinsky and he was not aware of any hard evidence that could be used against him. In Clinton?s memoirs he never admits to perjury. He admits to making misleading statements that were technically true. In the world of politics, a spin can be put on almost any statement to make it seem true, but Clinton said that he did not remember being alone with Lewinsky and that he only remembered giving her one gift and that he did not have sexual relations with her. In a political context Clinton might have been able to get away with these statements but in a court of law he could not. 155 By Clinton?s standards of political behavior, he told as much of the truth as he could possibly tell, especially when he was unaware of the possible evidence against him that would later come to light. If in Clinton?s mind he thought he could not be caught in a lie, than he had no reason to tell the entire truth as the RAM model would predict. So Clinton committed perjury and made the tenth worst presidential mistake in U.S. history. Rational Actor Model Table 8.3 below provides an outline of the four cases. Table 8.3: Summary of Presidential Mistake Cases Studied RAM explains decision? Game Theory model used Additional model used Additional model explain decision? Kennedy No Two person consecutive game None were appropriate but key personality traits were used instead Yes Johnson No Two person consecutive game GPM For the most part Nixon No Prisoner?s dilemma EAM Yes Clinton No Prisoner?s dilemma GPM Partially In all cases using another decision making model to determine the utility function for the RAM did not affect the basic premise of the RAM that decisions are based on the other players picking the most damaging alternative for the player being analyzed. What it seemed to do was either limit or reinterpret the alternatives available. Kennedy was unable to break the infiltrate alternative out from the execute alternative, and Johnson was unable to see pulling out of Vietnam as one of his alternatives. In the case of Nixon, 156 understanding his utility function showed that not losing the second term election was of more importance when he decided to cover-up the Watergate break-in than being impeached and staying out of jail later on. Finally, with Clinton, he actually followed the RAM by telling a limited form of the truth. If each of these presidents had not been governed by politics or elite power or a strong desire to win, then they could have analyzed their decisions using the RAM. Had they done so, they would not have made the mistakes they made. Kennedy would not have allowed the Cuban exiles to land on the Bay of Pigs, Johnson would have pulled out of Vietnam, Nixon would not have covered up the Watergate break-in, and Clinton would have admitted to another affair while everyone felt just as sorry for Hilary and she would have still become senator of New York. It might be useful to look at successful presidential decisions to see if those decisions were made following the RAM. If so, it would certainly be to a president?s advantage to evaluate all major decisions using game theory. The usefulness of game theory is in the structure it provides for making and predicting decisions. No other decision making model provides the structure and rules that game theory provides. The rules are based on the assumption that all players function according to the RAM. For a player to function according to the RAM, he or she must follow the following rules: 1. Behavior is goal directed 2. Goals are perceived to be in the actor?s self-interest 3. The actor will make a conscience choice between alternatives 157 4. The basic unit of measurement is the individual 5. The actor can rank outcomes in a priority order that will not easily change 6. The outcome with the higher priority will provide the most utility 7. Actors have enough information to make a determination (Monroe 1991, 4) The key to using game theory is determining how the player views his or her self- interests or utility function. Knowing the utility function can answer questions such as whether losing one?s job is a worse or better outcome than divorcing one?s spouse? Game theory can often be used without determining a specific utility function. The assumption in this case is that more money is better than less money, more power is better than less power, and longer jail time is worse than shorter jail time. But it cannot easily differentiate between the desirability of different outcomes. Without a defined utility function, game theory cannot determine how players evaluate complicated alternatives such as whether losing an election now is better or worse than being impeached later. When presidents, or anyone, have utility functions that force them to perceive situations and make decisions in a way that does not follow the RAM where a specific utility function is not known, they are not necessarily making decisions irrationally. It only means that to use game theory accurately at an individual level, a specific utility function is needed. If one understands a player?s utility function, there is a very good chance of predicting a player?s actions using game theory. This study tried to use well known decision models as a shortcut to determining a president?s utility function with a great 158 deal of success. If retrospectively the process can explain a decision, then there is a good chance it could predict one also. It cannot positively be determined if the process used in this study could have predicted the presidential mistakes before they happened. In the case of Kennedy, game theory could not have been used by an external observer like a political scientist because very little information on the Bay of Pigs game was available prior to the decision. Even an insider would have had trouble with the Bay of Pigs game because two game theory rules were violated. Kennedy was not given enough accurate information to effectively consider all the alternatives, and Kennedy could not make a conscious choice between two alternatives (execute and infiltration) because he was told that the same decision would activate both. Johnson?s decision also could not have been predicted using game theory because he did not pick his best worst case alternative unless looking weak would be a worse alternative to Johnson than plunging the U.S. into an extended war. Few would predict such an outcome using game theory even if it was understood that Johnson was an extremely political person. The context of the cold war and the domino theory also contributed to Johnson?s decision as well as the support of his advisors. But in the final analysis, it is difficult to believe game theory, even with a very political utility function, would have predicted Johnson?s escalation of the war, although it seems to do a good job of explaining the decision after the fact. Nixon?s prisoner?s dilemma also would be very difficult to predict but is possibly the case most likely to have been predicted by an insider. Unfortunately, an outsider would have never predicted a situation the public knew little about before hand. By 159 applying an elitist utility function, it would be possible to make a case for Nixon to agree to the cover up in order to prevent his worst case scenario of losing the election. As an elitist he would not consider that a president could ever be made to account for his crimes. Clinton?s prisoner?s dilemma would also be unpredictable by an outsider or even an insider except for the players themselves because of the lack of information. Even Clinton?s closest friends and advisors believed and defended him and felt betrayed upon learning the truth. Again game theory did a good job of explaining Clinton?s actions, but there would have been no way to predict Clinton?s decision ahead of time. The weakness in using this method is that to positively determine what model or models fit a person?s decision making requires a lot of information about the person and the situation. If that much information is gathered every time, the model may actually not be needed. As with Kennedy, who did not follow any of the defined models, all that was needed to still make an assessment was an accurate determination of the player?s primary driving forces. The three models applied here to the case study analysis may prove to be the most helpful if only limited information is known about a player. A player?s life history is not necessary to get enough of a sense of that player?s motives to determine which model is most appropriate. The availability of the facts of the situation may actually be the limiting factor in using game theory for predicting individual decisions rather than the utility function. The models could be used to provide the utility function needed to explain past decisions or possibly predict future ones but only if there is complete information about the situation, and that is not usually the case. 160 Conclusion This study began by calling attention to Gabriel Almond?s (Almond 1991, 48) question of whether sociology and psychology can be integrated into the rational actor model (RAM) without significantly changing the nature of the model, and if so, does this integration improve the model to the point where it can better predict political behavior. It appears that we are partially successful in answering this question by applying the RAM after using other models to determine the utility function of presidential mistakes. But by integrating these other models with the RAM to determine the RAM utility function have we changed the core theory of the RAM so that it can no longer be recognized as RAM as Almond feared? (Almond 1991, 48) In one way the answer is ?yes.? By relying on other models to determine the utility function of the RAM, we threaten the simplicity and elegance of the rational actor theory. The researcher can no longer merely count up number of years in prison or dollars lost or gained but must look into a person?s mind to determine how each person interprets his choices and their ramifications. This interpretation is far removed from the simple elegance of the RAM. But individuals are not simple, which means that the RAM will not always work when evaluating the actions of individual players unless one can discern their true utility function. The problem that lies before us is how to simply and easily determine that utility function. Models are one way, but there seems to be a very limited number of decision making models. Kennedy did not match any of the existing 161 models, and additional models do not exist to explain the motivations behind the current terrorist activity from Muslim extremists. The RAM works best when the utility function is correctly identified and there is complete information on the decision. Political scientists cannot control the level of information available to the public, but determination of the utility function can often be made by good psychological analysis of individuals or by using appropriate models. 162 REFERENCES Allison, Graham T., and Philip Zelihow. 1999. Essence of Decision, Explaining the Cuban Missile Crisis. New York: Addison-Wesley Educational Publishers. Almond, Gabriel A. 1991. ?Rational Choice Theory and the Social Sciences.? In The Economic Approach to Politics: A Cultural Reassessment of the Theory of Rational Action, ed. Kristen Renwick Monroe, 32-52. New York: Harper Collins Publishers. Arrow, Kenneth J. 1970. Social Choices and Individual Values. New Haven, CT and London: Yale University Press. Arrow, Kenneth J. 1977. ?Current Developments in the Theory of Social Choice.? Social Research, 44, No. 4: 607-622. Austen-Smith, David, and Jeffreys S. Banks. 1998. ?Social Choice Theory, Game Theory, and Positive Political Theory.? Annual Review of Political Science 1: 259-287. Baker, Peter. 2000. The Breach: Inside the Impeachment and Trial of William Jefferson Clinton. New York: Berkley Publishing Group. Barber, James D. 1992. The Presidential Character: Predicting Performance in the White House. Englewood Cliffs, N.J.: Prentice Hall. Barry, Brian, and Russell Hardin. 1982. ?Individual Actions and Collective Consequences.? In Rational Man and Irrational Society? An Introduction and Sourcebook, ed. Barry Brian and Russell Hardin, 19-37. London and New York: Sage Publications. Brams, Steven J. 2004. Game Theory and Politics. Mineola, NY: Dover Publications. Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor: University of Michigan Press. 163 Buchanan, James M. 1972. ?Toward Analysis of Closed Behavioral Systems.? In Theory of Public Choice, Political Applications of Economics, ed. James M. Buchanan and Robert D. Tollison, 11-23. Ann Arbor: University of Michigan Press. Caro, Robert. 2002. Master of the Senate: The Years of Lyndon Johnson. New York: Alfred A. Knopf. Clinton, Bill. 2005. My Life: The Presidential Years. New York: Alfred A. Knopf. Colodny, Len and Robert Gettlin. 1991. Silent Coup, the Removal of a President. New York: St. Martin?s Press. Copper, Terry. 1998. The Responsible Administrator: An Approach to Ethics for the Administrative Role. San Francisco, CA: Jossey-Bass. Davis, Morton D. 1983. Game Theory, A Nontechnical Introduction. Mineola, NY: Dover Publications. Dixit, Avinash K., and Barry J. Nalebuff. 1993. Thinking Strategically: The Competitive Edge in Business, Politics, and Everyday Life. New York: Norton and Company. Dixon, Norman. 1976. On the Psychology of Military Incompetence. New York: Basic Books, Inc. Publishers. Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row Publishers. Dye, Thomas R. 2001. Top Down Policy Making. New York: Chatham House Publishers. Ehrlichman, John. 1982. Witness to Power, the Nixon Years. New York: Pocket Books. Emery, Fred. 1994. Watergate, the Corruption of American Politics and the fall of Richard Nixon. New York: Simon and Schuster. Flyvbjerg, Bent. 2006. ?Five Misunderstandings About Case Study Research.? Qualitative Inquiry 12, No. 2: 219-245. Gelb, Leslie H. and Richard K. Betts. 1979. The Irony of Vietnam: The System Worked. Washington DC: The Brookings Institution. George, Alexander L., and Juliette L. George. 1998. Presidential Personality and Performance. Boulder, CO: Westview Press. Gleijeses, Piero. 1995. ?Ships in the Night: the CIA, the White House and the Bay of Pigs.? Journal of Latin American Studies 27, No. 1: 1-42. 164 Goffman, Erving. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Green, Donald P., and Ian Shapiro. 1994. Pathologies of Rational Choice Theory. New Haven, CT: Yale University Press. Gregg, Gary. 2006. Presidential Moments, February 18-19, 2006 conference. Louisville, KY: University of Louisville McConnell Center. Greenstein, Fred I. 2004. The Presidential Difference: Leadership Style from FDR to George W. Bush. Princeton, NJ: Princeton University Press. Haldeman, Harry R. 1978. The Ends of Power. New York: Dell. Halprerin, Morton. 1974. Bureaucratic Politics and Foreign Policy. Washington, DC: Brookings Institution Press. Hammond, Peter. 1975. ?Charity: Altruism or Cooperative Egotism?? In Altruism, Morality, and Economic Theory, ed. Edmund S. Phelps, 115-131. New York: Russell Sage Foundation. Hanyok, Robert J. 2000, 2001. ?Skunks, Bogies, Silent Hounds, and the Flying Fish: The Gulf of Tonkin Mystery, 2-4 August, 1964.? Cryptologic Quarterly Winter 2000/Spring 2001, 19/20: No. 4/1. Hess, Gary R. 2001. Presidential Decisions for War, Korea, Vietnam, and the Persian Gulf. Baltimore: The John Hopkins University Press. Higgins, Trumbull. 1987. The Perfect Failure, Kennedy, Eisenhower, and the CIA at the Bay of Pigs. New York: W.W. Norton. Holmes, Richard. 2001. The Oxford Companion to Military History. Oxford, U.K.: Oxford University Press. Holmes, Stephen. 1990. ?The Secret History of Self-Interest.? In Beyond Self Interest, edited by Jane Mansbridge, 267-286. Chicago: University of Chicago Press. Husain, Aiyaz. 2005. ?Covert Action and U.S. Cold War Strategy in Cuba, 1961-1962.? Cold War History 5, No. 1: 23-53. Isikoff, Michael. 2000. Uncovering Clinton: A Reporter?s Story. New York: Random House, Inc. Janis, Irving. 1972. Victims of Groupthink. Boston: Houghton Mifflin Company. 165 Jones, Bryan D. 1994. Reconceiving Decision-Making in Democratic Politics, Attention, Choice, and Public Policy. Chicago: University of Chicago Press. Kaiser, David. 2000. American Tragedy: Kennedy, Johnson and the Origins of the Vietnam War. Cambridge, MA: Harvard University Press. Klein, Joe. 2002. The Natural: The Misunderstood Presidency of Bill Clinton. New York: Broadway Books. Kutler, Stanley. 1990. The Wars of Watergate, the Last Crisis of Richard Nixon. New York: W.W. Norton & Company. Lammers, William W., and Genovese, Michael A. 2000. The Presidency and Domestic Policy: Comparing Leadership Styles, FDR to Clinton. Washington, DC: CQ Press. Lazo, Mario. 1968. Dagger in the Heart, American Policy Failure in Cuba. New York: Twin Circle Publishing. Levitt, Steven D., and Stephen J. Dubner. 2005. Freakonomics: A rogue economist explores the hidden side of everything. New York: HarperCollins Publishers. Liddy, G. Gordon. 1980. Will, the Autobiography of G. Gordon Liddy. New York: St. Martin?s Paperbacks. Logevall, Fredrik. 1999. Choosing War, The Lost Chance for Peace and the Escalation of War in Vietnam. Berkeley: University of California Press. Magruder, Jeb S. 1974. An American Life: One Man?s Road to Watergate. New York: Atheneum. Mansbridge, Jane J. 1990. ?Expanding the Range of Formal Modeling.? In Beyond Self Interest, ed. Jane Mansbridge, 254-263. Chicago: University of Chicago Press. Maslow, Abraham. 1943. ?A Theory of Human Motivation.? Psychological Review 50: 370-396. Mills, C. Wright. 1956. The Power Elite. New York: Oxford University Press. Monroe, Kristen Renwick. 1991. ?The Theory of Rational Action: Its Origins and Usefulness for Political Science.? In The Economic Approach to Politics, A critical Reassessment of the Theory of Rational Action, ed. Kristen Renwick Monroe, 1-31. New York: HarperCollins Publishers. Mueller, Dennis C. 2003. Public Choice III. New York: Cambridge University Press. 166 Muzzio, Douglas. 1982. Watergate Games, Strategies, Choices, Outcomes. New York: New York University Press. Neustadt, Richard E., and Ernest R. May. 1986. Thinking in Time, the Uses of History for Decision Makers. New York: Free Press. Neustadt, Richard E. 1991. Presidential Power and the Modern President: The Politics of Leadership from Roosevelt to Reagan. New York: Free Press. Nixon, Richard. 1978. The Memoirs of Richard Nixon. New York: Grosset & Dunlap. Olsen, Johan P. 2001. ?Garbage Cans, New Institutionalism, and the Study of Politics.? American Political Science Review 95, No. 1: 191-198. Olson, Keith. 2003. Watergate, the Presidential Scandal that Shook America. Lawrence, KS: University Press of Kansas. Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Boston: Harvard University Press. Ordeshook, Peter C. 1986. Game Theory and Political Theory: An Introduction. Cambridge: Cambridge University Press. O?Sullivan, Elizabethann, and Gary Rassel. 1999. Research Methods for Public Administration. New York: Longman. Posner, Richard. 1999. An Affair of State: The Investigation, Impeachment, and Trial of President Clinton. Cambridge, MA: Harvard University Press. Reedy, George E. 1970. The Twilight of the Presidency. New York: New American Library. Schelling, Thomas C. 1980. The Strategy of Conflict. Cambridge: Harvard University Press. Schlesinger Jr., Arther M. 1965. A Thousand Days, John F. Kennedy in the White House. New York: Houghton Miffin Co. Shelling, Thomas C. 1960. The Strategy of Conflict. London: Oxford University Press. Simmons, William. 1971. ?The Vietnam Intervention, 1964-65,? The Limits of Coercive Dipomacy, Loas, Cuba, Vietnam. Boston: Little, Brown and Company. Simon, Herbert A. 1945. Administrative Behavior. New York: Free Press. 167 Stucky, Mary and Shannon Wabshall. 2000. ?Sex, Lies, and Presidential Leadership: Interpretations of the Office.? Presidential Studies Quarterly, 30, 3, 514-533. Summers, Harry G. 1999. The Vietnam War Almanac. Novato, CA: Presidio Press. Thompson, John. 2000. Political Scandal: Power and Visibility in the Media Age. Malden, MA: Blackwell Publishers, Inc. Truman, David. 1971. The Governmental Process: Political Interests and Public Opinion. New York: Alfred A. Knopf. Walsh, Lawrence E. 1993. Final Report of the Independent Counsel for Iran/Contra Matters. Washington, DC: U.S. Court of Appeals for the District of Columbia Circuit. Warner, Geoffrey. 2003. ?Review article Lyndon Johnson?s war? Part I: Escalation.? International Affairs 79, No. 4: 829-853. Wofford, Harris. 1980. Of Kennedys and Kings, Making Sense of the Sixties. New York: Farrar, Straus, Giroux. Wyden, Peter. 1979. Bay of Pigs, The Untold Story. New York: Simon and Schuster. Yin, Robert K. 2003. Case Study Research, Design and Methods. Thousand Oaks, CA: Sage Publications. 168 Appendix A Letter of Request to take the Presidential Mistakes Survey November XX, 2005 Dr. XXXXXXX University of XXXXXX University Town State XXXXX Dear Dr. XXXXXXXX As an expert on the presidency and the American political system, we have chosen you to take part in a survey we are undertaking at the McConnell Center. Of course you know the old adage that says we learn more from our failures than our triumphs. With the recent troubles that have befallen the Bush Administration, we thought it might be time to take a little look at the great mistakes that have been made in presidential history. In February, the Center will be hosting a conference on important moments in the history of the office and no doubt most of those discussions will focus on the positive events and decisions surrounding our chief executives. At that conference we plan to release the results of this survey as well, and very much hope to be able to include one or more comments from you. Please take a few minutes to complete this brief survey and return it to us by December 9. In return, we would like to credit you for your participation and send along a little commemorative item from the forthcoming Grand Opening of the McConnell Center?s new headquarters on the campus of the University of Louisville. I thank you for your time and effort and look forward to hearing from you soon. Sincerely, Gary L. Gregg II, Ph.D. Mitch McConnell Chair in Leadership 169 Great Presidential Mistakes -A Survey- Please take a moment to complete this survey and return it to the McConnell Center by January 16th, 2006. You have been particularly chosen for this survey; we ask that you do not pass it on to any other person. Please be as complete as possible in your explanations as they will be crucial in helping us use this survey to teach about leadership and the presidency. I personally look forward to your answers and thank you very much for your time. Gary L. Gregg II, Ph.D. 1. President: _________________________________________________ Mistake: _______________________________________________________________ Circle a number on the scale below to indicate the severity with which the President?s mistake affected the nation. 1--------2---------3---------4---------5---------6---------7---------8---------9---------10 Hardly Highly Affected Affected Explanation: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________ 2. President: _________________________________________________ Mistake: ________________________________________________________________ Circle a number on the scale below to indicate the severity with which the President?s mistake affected the nation. 1---------2---------3---------4---------5---------6---------7---------8---------9---------10 Hardly Highly Affected Affected 170 Explanation: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________ 3. President: _________________________________________________ Mistake: ________________________________________________________________ Circle a number on the scale below to indicate the severity with which the President?s mistake affected the nation. 1---------2---------3---------4---------5---------6---------7---------8---------9---------10 Hardly Highly Affected Affected Explanation: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________ 4. President: _________________________________________________ Mistake: ________________________________________________________________ Circle a number on the scale below to indicate the severity with which the President?s mistake affected the nation. 1---------2---------3---------4---------5---------6---------7---------8---------9---------10 Hardly Highly Affected Affected Explanation: 171 ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________ 5. President: _________________________________________________ Mistake: ________________________________________________________________ Circle a number on the scale below to indicate the severity with which the President?s mistake affected the nation. 1---------2---------3---------4---------5---------6---------7---------8---------9---------10 Hardly Highly Affected Affected Explanation: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________ May we have permission to quote you and list you as a participant in this study? Yes No Signature: ________________________________________________ Date: ___________ Thank you so much for your cooperation and assistance. We greatly appreciate your participation in this survey. Please return responses by January 16th, 2006 to: Rachel Mulloy McConnell Center for Political Leadership University of Louisville Louisville, KY 40292 rachel.mulloy@louisville.edu 172 Presidential Scholars Polled for Presidential Mistakes Survey Randall Adkins - University of Nebraska at Omaha Andrew Busch - Claremont McKenna College: Gary Gregg - University of Louisville Marc Landy - Boston College Michael Nelson - Rhodes College Richard M. Pious - Columbia University David F. Forte - Cleveland-Marshall College of Law Michael A. Genovese - Loyola Marymount University Paul Kengor - Grove City College Robert D. Loevy - Colorado College Forrest McDonald - University of Alabama Thomas Mackey - University of Louisville Bruce Miroff - SUNY Albany Bruce Nesmith - Coe College James Pfiffner - George Mason School of Public Policy William Richardson - the University of South Dakota Dwight Vick - the University of South Dakota Andrew J. Dowdle - University of Arkansas Betty Glad - University of South Carolina Gary D. Glenn - Northern Illinois University Fred l. Greenstein - Princeton University Stephen Hess - George Washington University 173 Michael Kazin - Georgetown University Charles E. Walcott - Virginia Tech Louis Fisher - Congressional Research Service Matthew J. Franck - Radford University Shirley Anne Warshaw - Gettysburg College Thomas G. West - University of Dallas Karen M. Hult - Virginia Polytechnic Institute & State University George Edwards - Texas A&M University Michael Les Benedict - Ohio State University Thomas Wolf - Indiana Southeast University Judith A. Best - SUNY at Cortland Ryan J. Barilleaux - Miami University 174 Results of McConnell Center, Great Presidential Mistakes Survey, 2005 Expert President Mistake Affect Adkins Jefferson Embargo Act of 1807 8 Adkins Lincoln War against the CSA 10 Adkins McKinley Spanish-American War 9 Adkins Taft 16th Amendment (Income Tax) 8 Adkins Roosevelt, Franklin Social Security 10 Arnold Madison Proposing declaration of war of 1812 against British None listed Arnold Jackson Destruction of the second bank of the U.S. None listed Arnold Pierce Support of the Kansas-Nebraska Act and the principle of popular sovereignty None listed Arnold Roosevelt, Franklin Barring entry to US of European Jews attempting to escape Hitler and then during WWII publicly ignoring clear evidence of genocide while conducting the war against Hitler. None listed Arnold Reagan The assumption that cutting taxes is an absolute good Baker Buchanan Bias Toward the South 10 Baker Bush, George W. Iraq War 10 Baker Pierce Supporting Kansas-Nebraska Act and Kansas Policy 8 Baker Johnson, Andrew Reconstruction Policies 10 Barilleaux Wilson Refusal to compromise with Senate regarding the Treaty of Versailles 8 Barilleaux Nixon Watergate cover-up 6 Barilleaux Clinton Monica 4 Barilleaux Johnson, Lyndon Trying to have both the Great Society and the Vietnam War 8 Barilleaux Roosevelt, Theodore The Bull Moose campaign of 1912 6 Benedict Johnson, Andrew Refusing to cooperate with Republican majority in Congress in Reconstruction 10 Benedict Bush, George W. Iraq War policy 8 Benedict Buchanan Failure to take a strong stand against secession 7 Benedict Johnson, Lyndon Escalating the Vietnam War 6 Benedict Wilson Failure to broaden base for League of Nations 6 Best Wilson Unwillingness to negotiate League of Nation's Treaty with Senate 8 Best Johnson, Lyndon 1965 Vietnam decision making 8 Best Roosevelt, Franklin 1938 Court Packing Plan 5 Best Hayes Removal of federal troops from South 9 175 Expert President Mistake Affect Best Roosevelt, Franklin Negotiations with Stalin at Yalta 10 Busch Harrison, William Giving a two-hour inaugural address in the rain None listed Busch Buchanan Adopting feckless view that secession was unconstitutional but he had no authority to prevent it 10 Busch Johnson, Lyndon Committing to Vietnam with a game- theory no win strategy 9 Busch Nixon Not accepting responsibility for Watergate early 8 Busch Eisenhower Appointing Earl Warren to the Supreme Court 6 Dickinson Clinton Lewinsky affair 5 Dickinson Reagan Iran-Contra Affair 4 Dickinson Nixon Watergate cover-up 8 Dickinson Johnson, Lyndon Escalation in Vietnam 9 Dickinson Kennedy Bay of Pigs 3 Dowdle Nixon Watergate cover-up 10 Dowdle Eisenhower Denial of downing of U-2 plane in 1960. 6 Dowdle Reagan Iran-Contra Scandal 7 Dowdle Wilson Failure to compromise over the Treaty of Versailles 7 Dowdle Clinton Failure to tell truth about Monica Lewinsky 5 Fisher Johnson, Lyndon Gulf of Tonkin Resolution of August 1966 10 Fisher Bush, George W. Iraq Resolution of October 2002 10 Fisher Truman Korea War 10 Genovese Nixon Watergate 9 Genovese Reagan Iran-Contra 9 Genovese Clinton Monicagate 7 Genovese Bush, George W. Iraq and trampling on the U.S. Constitution 9 Genovese Buchanan Failing to act prior to outbreak of Civil War 10 Glenn Lincoln Using military force to keep the union together 10 Greenstein Kennedy Bay of Pigs 10 Greenstein Roosevelt, Franklin Court packing 8 Greenstein Clinton Monica 6 Greenstein Jefferson Embargo of 1807 8 Greenstein Nixon Watergate 10 Kengor Carter Mishandling of Iran 10 Kengor Johnson, Lyndon Mishandling of Vietnam War 10 Kengor Roosevelt, Franklin Internment of Japanese Americans None listed Kengor Roosevelt, Franklin Keeping his Vice President (Truman) in the dark 1 Kengor Truman Loss of China to communism 10 Loevy Nixon Watergate 10 176 Expert President Mistake Affect Loevy Madison War of 1812 8 Loevy Wilson Ahead of his time with League of Nations 7 Loevy Clinton Personal character 5 McDonald Jefferson The Embargo 7.5 McDonald Wilson Substituting morality for national interest in foreign affairs 9 McDonald Johnson, Lyndon Refusing to admit that there are things the government cannot and should not do 9 McDonald Jackson Destruction of the bank 6.5 McDonald Carter His "malaise" speech 7 Nesmith Buchanan Inaction during secession crisis 10 Nesmith Johnson, Lyndon Escalating US involvement in the Vietnam War 10 Nesmith Reagan Tax cuts 8 Nesmith Hoover Signing Hawley-Smoot Tariff 7 Nesmith Clinton Health care plan 6 Pfiffner Bush, George W. Invasion of Iraq 9 Pfiffner Johnson, Lyndon Escalation of U.S. military action in Vietnam 8 Pfiffner Truman Allows Douglas McArthur to drive north of the 38th parallel in Korea 6 Pfiffner Kennedy Bay of Pigs invasion 5 Pfiffner Clinton Do not take Osama bin Laden seriously enough 6 Pfiffner Bush, George W. Do not take Osama bin Laden seriously enough 6 Pious Clinton White House scandals 5 Pious Madison Agreeing to go to war against Great Britain 9 Pious Nixon Authorizing Watergate and related crimes and covering them up 8 Pious Johnson, Lyndon Authorizing escalation of American commitments in Vietnam 10 Pious Bush, George W. The war in Iraq 10 Warshaw Eisenhower U-2 Spy plane 1960; failed to question CIA 6 Warshaw Kennedy Bay of Pigs; failed to thoroughly question CIA 8 Warshaw Reagan Encouraging his staff to circumvent Congress 7 Warshaw Nixon Encouraging the Watergate cover-up 10 Warshaw Bush, George W. Sending troops into Iraq to overthrow Hussein 10 Wolf Hayes Pledge in inaugural address not to run for a second term 10 Wolf Johnson, Andrew Misjudging the mood of the country and Congress in reconstruction 10 177 Expert President Mistake Affect Wolf Roosevelt, Franklin The so-called "Judicial Revolution of 1837" 8 Wolf Clinton Failure to understand the risk of Islamic extremism 8 Wolf Nixon Fear of internal subversion 10