By: Carlos Pedro Gonçalves
(The current article is also available in the blog
http://riskgovernancestrategy.blogspot.pt/2013/01/theory-of-games-of-strategy-relevance.html)
Is the theory of games of strategy effective for current (geo)strategic problems that countries face?
Typically, the theory of games of strategy, involves a definition of a collection of players (each player being assigned a game position), a collection of strategies available for each player, a collection of payoffs associated with each strategic configuration, and game strategic reasoning criteria (largely formalized by von Neumann and Morgenstern and by Nash (see references at the end)).
For this theory to be effective in dealing with current (geo)strategic problems, these problems must be approachable from the formal framework addressed above, through some system of correspondences between the abstract formal and the concrete systemic situation to which game theory is applied.
However, are actual (geo)strategic problems effectively addressed by the theory of games of strategy worked upon by von Neumann and Morgenstern and, later on, by Nash?
Any problem that can be addressed by a finite (known) collection of players with a finite (known) collection of strategies and argumentatively definable set of payoffs can indeed be approached in some sense by this theory, however, in some cases, reality is not so clear-cut.
For instance, in the last few days Kaspersky Lab has denounced in its website a major sophisticated cyber espionage operation named “Red October” (“Rocra”) that affected all continents and has apparently been active for the last five years focusing on diplomatic and government agencies of various countries around the world (with special focus on the Russian Federation, Kazakhstan, Azerbaijan, Belgium, India, Afghanistan and Armenia, which together formed about 62.5% of the number of infections in the list of countries with most infections, as per Kaspersky Lab’s data and report (The_Red_October_Campaign)).
The major issue faced by Kaspersky Lab is the identification of the source of the operation. Investigating the “clues” left in the attacks, Kaspersky Lab identified two possible sources: (a) an operation perpetrated by some country or countries; (b) information brokers that may have harvested information to be sold to the highest bidders (a commenter on Kaspersky Lab’s blog also proposed the hypothesis of diversionary tactics as a criticism to Kaspersky’s analysis of the spelling errors of Rocra modules, a comment that adds more possibilities to the analysis). The data gathered by Kaspersky Lab, cannot thus lead to any conclusion as to which source may be behind “Red October”.
From a game theory perspective, one is looking at a played strategy, but all one can really see at work is the tactical maneuvers and targets that it involved. Kaspersky has identified hundreds of victims and eight main categories of targeted data:
- Government;
- Diplomatic/embassies;
- Research institutions;
- Trade and commerce;
- Nuclear/energy research;
- Oil and gas companies;
- Aerospace;
- Military.
Although the data regarding the played strategy is known, the player or players are unknown and only conjecturable to the point already synthesized by Kaspersky and commenters, so that the actual intention, motivation and payoff structure are not known, for each possible strategic scenario Kaspersky has identified possible payoffs in its report, but beyond that, the collection of all players involved cannot be known.
Cyberspace maneuvers, including cyber warfare, and hacker attacks by unknown groups or individuals open up a new issue that was not present at the time of the development of game theory: one does not always know the players, and their motivations; in some instances, one can only draw scenarios and, for each scenario, draw a possible game theoretical structure, so that one is faced with multiple games that may have been played, or that may be played.
In the current (geo)strategic framework, where, in the moving game board, companies, banks, States, global institutions, groups, individuals, epidemics/pandemics and diseases, natural disasters, economic, fiscal and systemic crises play a role, how can the theory of games of strategy be effectively operationalized?
In current (geo)strategic problems, we are left not with a single game theoretical solution but only with conjecturable, provisory sets of scenarios, games and solutions to which may be assigned plausibility gradations, in strategic and tactical risk analysis efforts.
More so that the fluidity with which new data comes to light, both at a publicly available level and at a private level, may invalidate some or all of the conjectures and scenarios drawn: natural disasters, financial catastrophes, fiscal problems raising, for instance, the risk for social and political instability, with the potential to radically change the game board make applied game theoretical analysis, as a strategic analysis tool, a fluid endeavor, in need of permanent revision and check for corroboration.
The dynamical reality of (geo)strategic problems is in stark contrast to applications of game theory such as the ones defended by Bueno de Mesquita (see reference at the end), where the whole stakeholder set, interests and weights are computable. Strategic analysis and risk analysis do not always have such a “clean” setting.
Unstable, unpredictable, (co)evolving, (co)complexifying systemic situations, with the sudden formation of extreme risk and catastrophe points that involve changeable players and demand a permanent monitoring and a well-built systemic analysis, do not fall into the stable decisional framework that characterized the early applications of the theory of games of strategy and demand a renewed look at this theory, integrating it with evolutionary scenario analysis in new mathematical frameworks, developed to deal with the current (geo)strategic problems and global interconnected risk situations.
References:
- Von Neumann, J. (1928). “Zur Theorie der Gesellschaftsspiele”. Mathematische Annalen 100 (1): 295–320.
- Von Neumann, J. e O. Morgenstern [1944], (1972). “Theory of Games and Economic Behavior”. Princeton: Princeton University Press.
- Nash, John (1950a). “The Bargaining Problem”. Econometrica, Vol. 18, no. 2: 155–162.
- Nash, John (1950b). “Equilibrium Points in N-Person Games”. Proc. Natl. Acad. Sci. USA, 36(1): 48-49.
- Nash, John (1950c). “Non-Cooperative Games”. PhD Thesis.
- Nash, John (1950d). “Two-Person Cooperative Games”. RAND P-172, August 9,
http://www.rand.org/pubs/papers/2005/P172.pdf.
- Nash, John (1951). “Non-Cooperative Games”. The Annals of Mathematics, Second Series, Vol. 54, No. 2, pp. 286-295.
- Mesquita, Bruce Bueno de (2010). "Prediction - How to see and shape the future with game theory". Great Britain: Vintage Books.