Essence of Decision

Graham Allison and Philip Zelikow's book Essence of Decision is much more than what its subtitle "Explaining the Cuban Missile Crisis" promises. It's a case study in how to think about international events using mental models. Allison & Zelikow explore three methods at increasing levels of detail:

• Model I: The Rational Actor — treat each nation as a single "person", making choices to maximize its utility, balancing costs and risks against rewards
• Model II: Organizational Behavior — analyze entities within each nation (agencies, departments, branches of the military, etc.) and their disparate doctrines, capacities, routines, resources, and propensities
• Model III: Governmental Politics — look at individual players in positions of authority and their clashing goals, bargaining styles, shifting alliances, etc., while recognizing the incompleteness of their information and the likelihood of mistakes

The events of 1962, when the USSR moved nuclear missiles into Cuba and then withdrew them, are Allison & Zelikow's archetypal example. In the concluding chapter the authors suggest a checklist of questions to ponder for each technique:

• Actors:
  • What's the situation — in reality, and as perceived — that offers threats and opportunities to each actor involved?
  • What are the primary goals of each actor?
  • What are the options each has available?
  • What are the costs and benefits of each option?
  • What's the best choice in this situation?
• Behaviors:
  • What sub-components form each actor?
  • What capabilities and constraints do each of those components (and their propensities toward "standard operating procedures") impose on knowledge of the situation?
  • What capabilities and constraints do the components create in generating options for action?
  • What capabilities and constraints do the components establish for implementing the chosen course(s) of action?
• Politics:
  • Who are the players (or forces) that matter?
  • What factors shape each player's perceptions, preferences, and stands?
  • What factors govern each individual player's impact on the group decision?
  • What processes exist to combine individual perceptions, preferences, and stands to result in a collective decision?

Analogous analytic tools directly apply to countless situations in history, business, and even personal life!

And what's missing? Perhaps an explicit recognition of probability — rough odds associated with competing hypotheses and alternatives — with thoughts on how to estimate and update those Bayesian likelihoods, along with comments on quantitative tests of one's models and their implications. Cue "Expert Political Judgment" by Phil Tetlock and colleagues, a few decades later ...

^z - 2019-04-09