In a recent FP article, Francis Gavin and James Steinberg observed that historical analogies can prove an unreliable guide to present-day policy decisions, specifically with regard to the momentous decision facing the United States and its allies regarding whether to strike Iran. As if to prove Gavin and Steinberg's point, Fareed Zakaria seeks in a Washington Post column to marshal two historical analogies in defense of his view that Israel should not attack Iran, but rather should seek to "contain" a nuclear-armed Iran if necessary. In doing so, Zakaria provides little insight into the difficult decisions facing Israeli or American leaders, but instead provides an instructive example of the fallacies that Gavin and Steinberg warn against.

Zakaria's first error is to cherry-pick historical analogies which fit what is presumably a preconceived conclusion -- that attacking Iran would be a strategic error. To support his view, he cites Germany's ill-fated decision to invade France in 1914, and the United States' decision not to attack the USSR in the late 1940s.

There are two problems with this sort of cherry-picking. First, Zakaria chooses only those historical cases which support the case for non-intervention, and ignores other possible analogies which might undermine his view. Just as critics of a strike like Zakaria could point to the cases he mentions or others to demonstrate how an attack could fail or non-intervention could succeed, advocates of a strike can cite the failure to confront mounting German militarism in the 1930s to highlight the risks of passivity, or cases of successful military interventions to illustrate the benefits of action.

Second, as with most broad historical analogies, both of the events cited by Zakaria are problematic as comparisons to the current tensions between Israel and Iran. Indeed, they must be shoe-horned into service to Zakaria's thesis. For example, Zakaria focuses on one factor which contributed to the outbreak of World War I -- German concerns about Russian armaments and mobilization capacity -- and excludes the many other circumstances which precipitated that conflict. And in citing the success of the decision to maintain a policy of "containment" (which was adopted prior to Moscow's development of nuclear weapons) toward the USSR rather than go to war, he fails to mention that this success came at considerable cost -- the domination of Eastern Europe for decades by the Soviets, the proliferation of nuclear weapons, and several major and countless minor wars.

Zakaria's second error is to commit, as many who employ analogies do, the logical fallacy of post hoc, ergo propter hoc -- that is, to assume that because an outcome followed a decision, it was caused by that decision. It is difficult to know even in retrospect whether the course of events depended on a particular decision or were in fact independent of it. Historians may wonder whether any decision by the major powers in the summer of 1914 could have averted a war in Europe, just as present-day policymakers are concerned that the Middle East will be more conflict-prone in the future regardless of Israel's decision regarding Iran.

Also problematic is the question of counterfactuals -- that is, whether different decisions would have produced outcomes better or worse than those which actually occurred. Historians argue vehemently over such issues, whereas partisan policy analysts have the cynical tendency to argue that anything that went well did so because of decisions their party or leader made, and that things that went poorly were either fated to do so or were someone else's fault. In reality, policymaking is a world of maddening ambiguity, in which not only outcomes but even facts tend to be uncertain.

Policymakers can -- indeed, must -- learn from history, but not by employing facile analogies in the service of preordained conclusions. History can help us understand problems and put them in their proper context; it can offer up novel solutions or shed new light on a dilemma; and it can warn us of the pitfalls that attend any decision and perhaps teach us how to avoid them. Learning from history is a tricky business -- in studying history, a policymaker must take lessons from one context and determine how and whether they apply to a different situation and a different era. Not only do different historical cases frequently suggest contradictory conclusions, but even individual cases -- for example U.S. arming of Afghan mujahedeen in the 1980s or the U.S. rapprochement with Muammar Qaddafi in the 2000s -- can offer multiple lessons which pull one in different directions.

As Gavin and Steinberg assert, in the end history cannot tell us what to do. Resolving thorny policy problems requires not just historical analysis, but also regional and strategic expertise, personal experience, and sound judgment. But above all, it requires the courage and conviction to choose, amid great uncertainty, among risky options. This is the essence of policymaking.