CIA: Project Success Among the toughest problems in innovating new games is choosing how to represent the basic real world situation that is being simulated. Too often designers simply fail to confront these choices in a creative way. The choices involved are much more than selecting what grain hexagon mapsheet is most suitable to a given piece of geography, or what the scale of miles to spaces should be. In fact approaching the problem of game system that way imposes real limits on what you can design, not just in the sense that the result will be another cookie cutter game-though that is a real possibility-but that the simplistic approach restricts the range and kind of situations that can be gained. To illustrate the potential for creative systems thinking, today we shall postulate an inherently difficult situation and then consider how to game it. What I have in mind here is a CIA paramilitary operation that took place in Guatemala in 1953-54. Called Project Success (technically PB/SUCCESS, but the first part of that CIA "digraph" was for regional or functional identification and we shall ignore it), the covert operation overthrew the government of Guatemala, which the U.S. government feared was veering off toward communism, and certain American corporations feared as a threat to their landholdings and business practices. After almost five decades, in mid-May 2003 the CIA released a large selection of documents and records that for the first time reveal the inner workings of Project Success. The Guatemala operation makes for a good example, not because it is a CIA operation or a relatively recent historical subject, but because from a game design standpoint it poses real challenges. First, a brief outline of the history. Project Success received approval from the Eisenhower administration in August 1953. The CIA spent months in planning and conceived its ultimate operational scheme that December, when a senior officer proceeded to Nicaragua to head the field staff for Success. There were several components of the covert plan. A series of networks of assets inside Guatemala were to be recruited, some for spying, others to enlist members for a resistance movement that, at the right time would rise up against the government, still others for the purpose of political action, making ostensibly independent proposals that in reality were to benefit the CIA operation. In parallel to the political action agents Project Success would set up a clandestine radio station that purported to broadcast for the fictitious (but, the CIA hoped, soon to be real) resistance. Meanwhile the agency would create a secret air arm that could parachute supplies and weapons to the inside networks. Finally, CIA officers would separately recruit, arm, and train outside the country a force under a Guatemalan commander which, at the appropriate moment, would invade the country and touch off the national rebellion against the home government. The various preparations went through three stages until, in June 1954, the CIA initiated its final stage of active operations with the invasion force. (The hypothetical game we are considering would center only upon this final phase of Project Success.) On June 18 the invasion force went in, some 480 men in sabotage teams plus several "shock" forces. On June 27 the national government capitulated and its president went into exile. This sounds straightforward except that it is not. Early on in the covert operation there were several damaging leaks that could have revealed details of the plan to the Guatemalan government. Indeed the president received a set of reports of suspicious activity from a friendly foreign diplomat as early as January 1954. Meanwhile the antigovernment Guatemalan agents who were supposed to have all those resistance fighters ready to rise up were in fact fabricators with no such numbers of adherents. (One Guatemalan CIA agent claimed to have over 39,000 people ready to rise up by January 1954. When the invasion came there were no significant uprisings at all.) At the end of May one of the most important Project Success internal networks was broken up by government security forces and its radios and codes compromised. When the time for the invasion came, one of the CIA teams was simply arrested by local Salvadoran police who found the troops in their country. A "shock" team of over 100 men was routed by a government outpost manned by 30 soldiers. The main force was defeated in its own bid to capture another Guatemalan outpost. Meanwhile the Guatemalan government had received a shipment of arms from Eastern Europe that the CIA had failed to stop, and its effort to sabotage the rail line the Guatemalans had to use to move the weapons from an Atlantic coast port to the capital also failed. Finally, in contrast to the few hundred CIA "shock" force troops, the Guatemalan army had 6,000 men. The clear problem in this from a game design standpoint is that standard methods would afford our CIA player practically no chance at success. A hexagon map of any scale would be of little use-the game effectively ends at a point the CIA "shock" forces are just a couple of hexes into the Guatemalan government player's territory. A standard force representation, no matter how you cut it, also gives the CIA player a ratio of less than 1-12 against his opponent. Even worse, this is not a case where we can accord the CIA side exorbitant morale and qualitative advantages to help make up for their numerical inferiority. Their battlefield performance once the invasion began is far too poor to justify such a measure. Equally problematical, the CIA's anointed leader, a former Guatemalan army officer, was a very bad commander, ruling out any significant leadership bonuses. An area map done up for Guatemalan provinces makes a little more sense. Here the CIA player could be challenging political control over border provinces. This approach would also accommodate another feature that seems necessary: some sort of political scale that can be used to measure political loyalties. Such a scale could vary province by province or nationally, and could be used to determine recruitment of resistance forces (although none were mobilized in actuality, and absurdly nonexistent numbers were claimed in January 1954, that does not mean no such recruitment was possible). The area map approach would enable the designer to represent the tactical interaction of the frontier fighting that occurred in the Project Success invasion, at least to some degree. In addition, since this CIA covert operation is a political- military campaign, not a simply combat simulation, the area map approach would help us get away from some of the thinking that usually infects us when we sit down at a hexagon mapboard. What defeats the area map approach is time. In the typical political-military simulation in which this technique is used, time is measured in weeks, months, or even years. In Guatemala 1954 there are just nine days from the moment the first CIA raiders step across the border to when the Guatemalan president resigns. Even if you take the starting point as when the CIA force sets off from base camp (June 15), and end the game at the point where the American covert agency calls off its campaign after the victory (July 2), the elapsed time is just seventeen days. That would be barely enough to make an unobstructed route march into the country, not at all enough to portray a political process by which loyalties shift from one side to the other. Were we to design the game so that this nevertheless was the major simulation technique, the process would have to be so artificial that the game would attain little credibility. How did the CIA win in Project Success? The evidence shows that the main drivers were the air campaign and the propaganda operation. On the air campaign, transport planes were supplemented with a handful of fighter aircraft just before the onset of active operations. However, the force size (three fighters at start, three more later, with one plane lost in action and one in a crash landing) is such that air lacks real military power. In addition, the U.S. provided only a handful of bombs to the CIA force (20 120kg and 20 250 kg munitions), strengthening the conclusion that these assets were really for demonstration purposes. Ultimately the air campaign should be viewed as contributing to political shifts that were also the direct target of the psychological warfare campaign. The radio station the CIA set up constantly harped on antigovernment themes, CIA agents covered Guatemala City with graffiti suggesting a mass opposition movement, and U.S. diplomatic moves and overt propaganda (such as Voice of America broadcasts) all contributed to this campaign. The totality of these efforts did not ultimately undermine the loyalty of the Guatemalan army to its government, as the army's resistance in the border fighting shows. CIA hopes in this regard were dashed. But at some point President Jacobo Arbenz of Guatemala (who had been elected fairly in a democratic election) lost his nerve and fled to Mexico. Thus for our hypothetical game the psychological developments are the kev ones. The optimum mechanism to portray this series of events in a game would be card play. The design would have to play to Arbenzs loss of nerve, and would use the kind of political loyalty measure discussed earlier. I would recommend cards for loyalty (in varying denominations). There would also be dummy cards, ones for propaganda (portraying, and valuing, particular themes and methods), for military forces, cards for specific events in the Project Success campaign (first air drop, first bombing, unauthorized air mission that embarrasses CIA, invasion, army mobilization, new recruiting, agent net arrested, security breach, arrival of arms shipment, CIA propaganda exploitation of communist arms, military defeat for CIA, etc). There are plenty of these events that can be researched in the record of Project Success. A card process would free the design from the constraint of the board, and the narrow time frame of June 18 to 27. The game could run in untimed turns and portray events from the onset of Project Success forwards. At the beginning of the game the Guatemalan player could lay down a number of political loyalty cards (depending on his card draw), and the CIA player perhaps just one. Propaganda and Action cards would be playable by only one player (whether there should be separate player decks or one combined deck, and whether a player could deny cards from the opponent by holding them in his own hand would be worked out from testing) After that, through play of propaganda or action cards, the CIA player could steal loyalty cards from the Guatemalan government, and that player might be able to occasionally reinforce his strength showing on the table with new cards from the deck. Some actions both players could take would cancel the effects of actions by the opponent. At the end of play the player with the most loyalty and military force showing would be the winner. That is the end of our game design road. The point here is not that Project Success can or should be designed as a game. The point is that what we can portray in games is directly influenced by what design techniques we adopt. Here we do a game where a side that has a reasonable chance of winning is inferior by 1- 12 in strength, has poor morale, execrable leadership, a marginal air capability, and so on, while a side with commensurately greater advantages has a reasonable chance to lose. With a mapboard and counter game that could not have been done. Design away, but never fail to apply imagination from the very first concept for your game! Back to Table of Contents -- Against the Odds vol. 2 no. 1 Back to Against the Odds List of Issues Back to MagWeb Magazine List © Copyright 2003 by LPS. This article appears in MagWeb.com (Magazine Web) on the Internet World Wide Web. Other articles from military history and related magazines are available at http://www.magweb.com * Buy this back issue or subscribe to Against the Odds direct from LPS.