Intuitively, it appears somewhat surprising that states’ foreign policies are fairly rational, for why should we expect sub-state politics, bargaining and decision-procedures to translate into “rational” decisions and policies. Of course, one might take issues with the premise that decisions taken by, or policies pursued by, states are reasonably (that is, on average, in the long run, and so on) rational. Then it is perhaps still surprising that foreign policies are not much more “all over the place”, namely substantially more irrational than they are. Admittedly, this is all somewhat, or even more than somewhat, subjective. So let’s try to disentangle these issues – conceptually.
Rationality is a slippery, ambiguous concept. Rationality can be defined as a means-ends relationship. A decision is irrational if it seeks ends that are not achievable given available means. A decision is also irrational if it has verifiably suffered from cognitive to affective biases. A policy is irrational if it is maintained in spite of evidence strongly supporting the need for a change in policy and such a policy is available at an acceptable cost. Inevitably, evaluating individual decisions will often involve a certain degree of subjective judgment, even if one agrees on all the facts.
Rationality should not be evaluated in terms of the success or failure of a decision or policy. A decision or policy that ends in failure need not have been irrational. A policy that turns out to be successful may have been deeply irrational. Virtually all decisions rely on assumptions and presumptions, and decision-makers often suffer from misperception (which may or may not be due to systemic biases and therefore irrational). Neither should the question be whether things could have gone differently (they almost always could have), but whether a better decision could have been taken – not in hindsight, but in view of what was known (or should have been known or could have been known) at the time. Again, this leaves no doubt significant room for disagreement when it comes to evaluating the rationality of a decision.
For example, the Battle of Midway could have gone differently than it did had the Japanese navy had a different carrier doctrine, invested more in intelligence and reconnaissance, or had a Japanese patrol plane had not failed to spot US carriers due to cloud coverage, or had the task force been less confident that their attack would come as a complete surprise to the US. This raises an important question about an individual decision and the context in which a decision is taken. The individual decision may have been irrational, but the context within which irrational decisions was taken may itself have been due to “irrational” decisions earlier on. These decisions, in turn, may have been due to political or bureaucratic comprises. flaws in the decision-making process or cognitive biases. Changing any of these decisions may plausibly suggest that a successful outcome might have been achieved. But hindsight bias needs to be avoided. The question about “how far back” an irrational decision reaches is nonetheless an important one and deserves greater exploration.
Decisions are determined by personal, bureaucratic (organizational process) and political (governmental politics) interests, and they are affected by decision-making procedures. Graham Allison (1971) famously made this point famously in his study of the Cuban missile crisis. This may translate into irrational policies (however ascertained) due to the need for political or bureaucratic compromises or the failure to correct for cognitive and affective biases. It may also lead to important issues not being raised as well as important questions not being asked. Japan arguably lacked a grand strategy because foreign policy and military intervention were the consequences of bureaucratic compromises rather than a high-level, political-strategic decision-making process.
The reality is that many of a state’s decisions (and bureaucratic, non-decisions) do not have dramatic consequences, and the costs of irrational decisions are often limited, or such decisions can be reversed once they begin to reveal their “irrationality”. As Jon Maynard Keynes quipped: “When the facts change, I change my mind. What do you do, sir?”. At the highest decision-making level, this requires deciders to ignore the sunk cost effect. At the policy implementation level, bureaucracies (often, not always) need to be “learning organizations” in order to adjust (as opposed to simply reverse) policies. Battles may not be refought, but tactics can be adjusted. So can strategy, in principle. Put differently, often there is room for maneuver, adjustment and recovery. Not always. In the face of German military aggression in 1939/ 40. Britain managed to recover, while France could not.
Allison’s organizational and governmental politics model may have been somewhat oversold (Krasner 1972). But both models provide heuristically valuable frameworks for analysis. And they no doubt help explain curious and important facts, such as why the Soviets did camouflage the construction of their missile bases in Cuba, or even why German and Japanese intelligence was sub-par compared to their respective operational skills. (Cultural and geo-strategic accounts may also help explain the latter.) But high-level decision can often (but not always) be adjusted or reversed, and (often, not always) the bureaucracies prove sufficiently adept at adjusting or reversing course.
Admittedly, these are broad, largely conceptual statements, supported by (at best) anecdotal evidence …. This handful of paragraphs is less an argument than food for thought and a call to revisit aforementioned “intellectually neglected perspectives”. The intention is to revisit this question and the concomitant analytical framework in greater detail at some point.