This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected. Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies. Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue. The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.
War games, popularized by last summer’s blockbuster movie of the same name, right, are extremely useful tools for studying warfare but are less appropriate, perhaps even misleading, for investigating some questions. Our understanding of war games and what we can learn from them lag behind their growing popularity.
A war game consists of three elements: a game facility, a scenario, and people to play the game. For a simple tabletop box game, the facility might consist of a room, table, chairs, and the game board and pieces. The scenario is that military conflict situation which the game depicts. Taking a concrete example, Table 1 lists the makeup of these elements for the war games at the Naval War College’s Center for War Gaming (CWG). The current CWG facility is large and includes many different command centers and associated communication links, a computer for keeping track of unit positions and status, and models to resolve detections, engagements, and damage assessment.
The game scenario is spelled out in an operations order (OpOrd), the primary document for the game; it contains the political backdrop, order of battle, current intelligence, operations area, and force missions. Game players assume the role of task force, task group, task unit, and task element commanders depending upon which commands are being played. They usually have an associated staff. Umpires, with the help of the computer, act as subordinate commanders, to execute orders from commands being played, as well as operate subordinate forces.
Figure 1 outlines the flow of information in a CWG war game. Those playing the role of force commanders occupy “command centers” away from the game floor. Each command center is a separate room a with computer display terminal, plotting boards, status boards, telephones, | and other communications gear. Commanders act on the f tactical picture which they develop from information from 8 these voice, hard copy, and computer circuits.
The game controllers, umpires, and higher authorities * strive to provide a realistic flow of information to the play- I
82
w Mystique
Geographic region Time
Commanders
Staffs
Forces and capabilities Initial positions and posture Missions and directives Constraining rules
Game
controllers
Umpires
Higher
authority
Game Sponsor
of
A war game commander also is denied the option
Table 1 Elements of a War Game
Facilities Scenario People
Command centers Communications circuits Computers
Umpire tables
Models
Force movement bookkeeping Detections Engagement resolution
Damage assessment
___ 7T~V________ -yrf 'K | ||
Blue Umpire Tables |
| • • • Orange Umpire Tables |
| ||
|
'i
Game Director
ers. Higher authorities try to ensure that the theater political context and the reaction from the rest of the world lead the game in the desired direction.
The umpires control the individual force units, keeping track of positions, understanding the tactics and procedures their forces have been directed to employ, and reporting information (radar and sonar detections, engagements, etc.) which their units would logically have available to them. This last point is crucial to the game. With 20 or more umpires passing contacts, force status, and other information to the command centers, it is imperative that the information be realistic and that the umpires coordinate both among themselves and with the higher authorities.
War games offer several kinds of learning experiences- One kind centers on a player’s direct involvement in g111112 activities which generates a collection of personal expert' ences. The usefulness of these experiences depends upon the war game’s degree of realism. War games, like exercises, are an imperfect image of war, since these ships do not steam around, and these planes do not drop bombs or fire missiles. But some war game activities do closely approximate actual war fighting at sea.
Consider as an example an actual battle group (B® commander with assigned forces and a mission. He and his staff must plan an operation to carry out that mission and then oversee the execution of the plan. During planning, they analyze the problems confronting them by assessing enemy capabilities, identifying possible enemy courses of action and own courses of action, and decide upon a concept of operations. A thorough analysis requires understanding the environmental and political factors, translating these factors into operating constraints- and devising ways to exploit these constraints. In both written and oral communications, the BG commander must convey to his subordinates his battle philosophy- identifying those decisions he reserves for himself and those decisions delegated to them. He must identify me information required to make informed tactical decisions- and how to keep track of and evaluate that information. ln the heat of battle, he becomes a crisis manager, weighing staff advice, assessing tactical situations, monitoring hattie progress, and amending his initial plans and force disposition accordingly.
Most of these real-life activities also occur during CW<J war games, but the environment is different. First, some command center facilities are at present rudimentary compared with those on board modem flagships, which makcs it more difficult to maintain a picture of the tactical situation in the war game than it would be during a real opera" tion at sea. The paucity of status boards and circuit talked in a war game is an example; display facilities can also he lacking. On the other hand, CWG’s new multi-milli°n dollar Naval Warfare Gaming System’s display console and access to status information may actually be supert°j to what is available on board ships at sea, at least unh Aegis ships are around in significant numbers.
monitoring battle progress by selectively monitoring tad1" cal warfare coordination or reporting circuits, which is a cornerstone of modem command and control methods- Both the number of circuits and the data transfer rate can be deficient. On the other hand, the reliability of individual communication circuits is much better in the war gade than in the real world. Further, many of the communication links could be lost in a hot war, yet this aspect 01 warfare is seldom, if ever, played during a war game- Therefore, while communication circuits are much more plentiful in the real world, the few which are used in a v*'ar game are unrealistically invulnerable. The lowest command level being played cannot directly confer with a stilj lower level commander. Orders to units are passed through umpires who act instead as a collection of individual platform commanders. Hence, the superior-subord1'
a e relations that a military command structure can en- nce is an advantage that cannot be exploited in a war & me. Finally, the war game environment lacks the nd-the-clock strain and stress of at-sea operations. the016 Players can retire to their rooms in comfort, relax on e tennis courts, and get away from the war game; the oc|» guts, gore, and fear are missing. There are a host of actical issues with which the war game BG commander a not concern himself: personnel problems, routine Ports, normal ships material conditions, safety during gnt operations, etc. Personality clashes arise among the f - yers> and battle damage does occur, but the problems short of those a BG commander experiences, n balance, many of a BG commander’s real-world I Rational activities (summarized in Table 2) are simu- . reasonably well in war games. Experiences stem- lng from these activities faithfully reproduce likely ex- [i,riences from a similar real-world situation. A war game en *s a good vehicle for learning about the BG com- Sk^der’s j°b, or for a BG commander to practice needed
list in Table 2 also applies to almost any level of ^°nimand, not just the BG commander, so long as the a°mniander has some tactical latitude, there is someone to ^ct as his subordinate (even if only an umpire), and the attle problem is complex enough for geography, environ- ent, and politics to have some influence. The detail will ^ePend on the command level, but the generic activities 11 remain much the same.
e ^n°ther kind of learning from war games is based upon l1- amination of the sequence of naval events, or the game •story, along one of two lines. First, one might delve into e underlying logic of a battle episode to determine why tje §ame events led in a certain direction. This investiga- °u would treat questions concerning the decisions which ere made, the important characteristics of the scenario, e rationale for each side’s actions, and the result of dif- rent actions.
ae second line of attack treats the game history as a e$UrC£ sc*ent'f'c evidence on matters of research inter’ such as tactics, employment of new platforms or eapon systems, and certain organizational or procedural $isUS ^nvest*§ations of the first type require sound analy- techniques of good analytical historians or investiga- e reporters, while investigations of the second type °re resemble scientific inquiry. s ^ comparison of real life with the game history yields ^ eral major differences. First, a war game’s scenario is e designer’s conception of the ways in which events u <a lead to a naval confrontation and the general shape mat confrontation. While the aspirations and motiva- ns attributed to countries may be believable, and the jj. umng battle problems seem plausible, one is never sure ue scenario is truly realistic. Second, forces do not Ually operate in a war game. Instead, these operations e simulated either using a computer or manually by drr>e control personnel. Hence, the multitude of operating °blems which beset naval forces at sea such as personnel °mems, material problems, manpower skill levels, and
unexpected calamities play almost no role in a war game. Third, in a war game, detections, engagements, and damage are the result of applying models to a tactical situation resulting from player decisions. Good models can capture only the most important factors affecting these evolutions and therefore can only approximate reality. Some of the models could be poor approximations, which would inevitably lead to questionable episodes in the game history.
The area in which the war game does seem to closely mirror a real battle is in the decision-making process. Many of the necessary analytical steps are present in a war game: monitoring the situation, assessing tactical status, making a decision, and living with the results of the decision. One drawback, however, is the context of the decision-making process: players know that the play is all make-believe. Ships do not explode, people do not die, and the balance of world power does not hinge upon the game’s outcome. Since not as much is riding on the decisions, players might be more aggressive, flamboyant, or in some cases, overly cautious, than they would be in the real world. Thus, while the decision process looks reasonably genuine, the decisions may not ring true.
Given these “artificialities” inherent in war games, the game history provides only tenuous evidence for some kinds of questions. Consider, for example, the question of judging the worth of a tactic. A tactic is usually tried out a number of times in several candidate situations or scenarios. To conclude that the tactic is a good one requires a pattern of successful results correlated with use of the tac-
Table 2 A Battle Group Commander’s Operational Activities Tactical
Force selection and employment Integration of platforms to accomplish a task Battle group tactical decision-making Exploitation of platform/system limitations Rapid assessment of tactical situations Real-time revision of operational plans in response to a changing situation
Command
Delegation of authority
Articulation of battle philosophy, directives, and orders Establishing information requirements for decisionmaking
Devising effective ways to display and evaluate information Assessment of advice Crisis leadership
Scenario
Exploitation of geography
Exploitation of environment
Exploitation of international political relations
tic instead of some other factor. A single war game would not normally provide the needed pattern of trials, but more importantly, success in the war game does not necessarily mean success in the real world. The “success” of a tactic in a war game is too strongly dependent upon the fidelity of the detection, engagement, and damage assessment models, and this causes problems. If the models capture the essence of naval battles, they may well be insensitive to the factors which a new tactic postulates as crucial. The new tactic may be doomed to failure if tested by a method relying on models which are not detailed enough to “recognize” these factors. The models may either be biased against, or be otherwise inappropriate for, testing the tactic. To derive reliable conclusions about a tactic’s success demands reliable, sensitive models. But if good models are available, one wonders why the war game play would be needed at all.
With the proper pattern of trials established beforehand and with reliable models, one ought to be able to generate as much evidence as one gets in a war game by cranking through the models the requisite number of times for each situation in the privacy of an office. While it is true that a particular tactic may undergo substantially closer scrutiny in a war game than in someone’s office, a war game will not substantially help judge the tactic’s chance of success. Nor will a war game furnish much information practical matters of execution such as ease of implementation °r integrability with other standard operating procedures. S° if the models are good, a game episode furnishes lime evidence for rigorous tactical evaluation which cannot h® gotten in an easier way. But if the underlying models either are of unknown or poor quality, then the game ep1' sode may even end up providing misleading evidence-
For many of the same reasons, definitive examination of new weapon systems or new platforms in war games18 equally unpromising. However, a war game’s invigorating atmosphere may stimulate players to explore new ways to employ the platform or system. Players are forced to thin about hard military problems in a stressful environment, would not be surprising if new ideas popped up during d*e game. It would be more surprising if they did not. Extraction of these innovative ideas is worthwhile, but using v“ir games to “test” tactics, new systems, or new platform8 seems ill-advised.
The underlying models also pervade evaluation of spe cific decisions made during the game. If the outcome of 3 decision is a factor in the judgment process, then the models’ reliability is an important factor since they determine outcomes. Unless the models are known to be accurate- these judgments must be made with caution. A more rel<3' ble method of judging a decision is to base it primarily °n the conditions surrounding the decision. The question really is: “Given the information available to the dec*' sion-maker at the time he made the decision, does it seem to have been a good one?’ ’ The answer requires examining the information available to the decision-maker, the tactical situation derived from that information, the mission- and the directives and guidance he was operating underAside from the outcome of the action, if one can make 3 convincing case that the decision was poorly thought oU<- then the judgment would not rest on the fidelity of th6
underlying models. However, these kinds of arguments rarely conclusive. Finally, even in the case where the , s are known to be accurate, an outcome may be the yu * drawing a rare random number. Hence, the unfain th 6 .outcome *s actually a rare outcome for the action at situation. So, even in this case, judgments concern- § game decisions must be made cautiously.
■ "e °nly remaining questions are those which might be C ,ea scenario issues”—issues which concern the flow ^ attle and the battle as a whole. For example, it would Use^ui to identify and remember shrewd own courses of to °i?' °r enemy courses of action which pose particularly in th Pro^ems *n striving to accomplish a given mission the 6 Scenar*°- ^ would also be useful to know the factors eev°lution of the scenario and its outcome hinge upon, for ’ §eo8raPhy or water conditions, the presence of ^ c.es with a certain weapon system or other capability, tjr lmP°rtance of securing a tactical objective, or the iden- jj- Catlon of an especially critical decision. Often the game
0 st0ry can be a source of answers without undue reliance Setae character of the underlying models. Any reasonable
°f models would likely yield the same answers, in > °r conclus've answers, the influence of any confound- 8 execution or operating difficulties plus the fog of war st be accounted for; this still requires at-sea operations
1 a real forces. Yet for scenario issues, the war game ^ comes an ideal forum for “organized brain storming.”
ach set of players plunges into a military problem and is rrced to make decisions. This is a much more fertile envi- °nment for seeking out solutions than the sterile office or conference room.
u Effective learning from a war game depends strongly Ron how closely the war game experiences, activities, in J*16 resulting game history resemble similar evolutions c | real world. Unfortunately, many game artificialities ^ or player experiences and activities, plus the course of (jele.events. The models simulating force movement, ection, engagement, and damage assessment also induce the game’s outcome. One cannot blithely treat the 111 c experiences as if they were the same as correspond- 8 experiences in the real world. This is not a fatal indict- nt of war games since judicious interpretation of war oie experiences and aggressive involvement in the war oie can still make war gaming a tremendously fruitful j^Perience. And in peacetime, war games, together with ct exercises, constitute our most realistic source of such Periences and scenario information.
War game activities related to the decision-making of oie commanders are reasonably close to their real-world . *)unterpart. The type of thinking and the mental skills v°lved are exactly those required in battle. One must ^asP the military problem, analyze it, select and deploy f()rces, develop and maintain a tactical picture, control *s, assess tactical situations, and act decisively, viame realism becomes more questionable when one Osiders the circumstances of the decision-making: the . of information available to the decision-maker, the jolablc facilities to process the information, and the fi- uty of information How to and from the force commander. These contextual factors color the decision-making experience.
A second group of problems arises if one attempts to use the game history as evidence in arguments. The phrase, “we tested it in the war game,” is becoming more popular, but perhaps not justifiably so. The most obvious hurdle to overcome is determining the reliability of the underlying battle simulation models. The outcome of a battle evolution in the game relies heavily on these models, and bad models mean unreliable outcomes. The outcome of the war game “test” should be viewed skeptically until doubts about the models can be resolved. The same holds true for second guessing tactical decisions made during the course of the game.
A group of issues which a war game can shed light on are those related to our own courses of action, enemy courses of action, and concepts of operations. Courses of action, concepts of operations, key factors upon which the tide of battle depends, and crucial decision points undergo close scrutiny by many different players and from many different perspectives. While the result of this scrutiny is seldom definitive, the war game will still enhance understanding of the basic naval problem.
The fundamental usefulness of a war game is to provide decision-making opportunities for the players and to provide a forum for systematic brainstorming for all those involved. Making decisions and struggling with scenario issues are cerebral endeavors. Only the commanders get the opportunity to actually make the decisions, but all have the opportunity to examine the decision from their game perspective, be it game sponsor, umpire, higher authority, staff player, or observer. One must be immersed in the game and “think through” the problem. Benefit, too, will come from post-game discussions during which participants understand how seemingly sound decisions at the time appear ill-conceived from other perspectives. Lack of involvement on the part of players, controllers, or observers will make the war-gaming experience a waste of time.
While 1 have discussed CWG’s games throughout this article, this is a common sense position from which to enter almost any war game. All war games, whether run by the Joint Chiefs of Staff, other elements of the Naval Education and Training Command, or the new tactical training groups, approximate CWG’s games.
Each war game simulates some activities well and some poorly, and each is dependent upon models, be they crude or sophisticated. War game experiences are unquestionably of value if we but use a bit of common sense; there is no magic.
Mr. Thompson received a master’s degree in statistics from the University of Rochester and is a PhD candidate in economics at the University of Virginia. He has been a member of the Operations Evaluation Group (OEG) of the Center tor Naval Analyses since 1976. Over the past three years, he has participated in, designed, conducted, and analyzed numerous war games as Professor of Naval Operations at the Naval War College and as OEG Representative to Commander Sixth Fleet. He is currently the OEG representative to CinCUSNavEur.
___________ Second Honorable Mention___________
Education and Training Minicontest