On March 19, the New York Times described a classified U.S. Central Command war game conducted this month that simulated the outcome of an Israeli attack on Iran. According to U.S. officials who discussed the results with the newspaper, the game "forecasts that the [Israeli] strike would lead to a wider regional war, which could draw in the United States and leave hundreds of Americans dead." Marine Gen. James Mattis, commander of Central Command, found the outcome "particularly troubling" because an Israeli first strike would have "dire consequences across the region and for United States forces there."
The article, with its discussion of "dire consequences," is one more indication of the gap between the Israeli government's calculations concerning Iran and those of the U.S. government. Why that analytical gap exists should be of interest to policymakers. The military's conclusion that U.S. forces in the Persian Gulf region could suffer hundreds of deaths following an Israeli strike could be an indication that U.S. commanders and policymakers have not adequately prepared for such a scenario. But perhaps most important, we should examine what goals U.S. officials had in mind when they leaked the results of the supposedly secret war game to the New York Times.
According to the article, the two-week Central Command war game, called Internal Look, was specifically designed to test internal military communications and coordination among battle staffs in the Pentagon, Central Command headquarters in Tampa, and field units in the Persian Gulf. According to the scenario, Iran would conclude that the United States was an Israeli partner and therefore U.S. military forces in the Gulf were complicit in the Israeli first strike. The simulation had Iranian anti-ship missiles strike a U.S. warship killing hundreds of sailors. The United States then retaliated with its own strikes on Iranian nuclear facilities.
This simulation appears to differ sharply from Israeli expectations. According to Jeffrey Goldberg's reporting at Bloomberg, Israeli officials believe Iran will not target U.S. ships or facilities in the region because it would hardly be in Iran's interest to bring Central Command's military power into the conflict (a point I made in a recent column). Indeed, according to Goldberg, Israeli policymakers believe that if Israel's strikes are limited to a handful of nuclear targets away from urban areas, Iran might actually downplay the severity or cover up the damage, as Syria did after the 2007 Israeli strike on its nuclear reactor.
Since Internal Look was designed to give U.S. military global command and control systems a workout, it would not help commanders achieve that objective if the scenario didn't escalate up to high-intensity combat action. Requiring the scenario to do that is completely different than having the war game objectively conclude that such escalation is the most likely outcome -- a conclusion Israeli planners presumably don't share. If Internal Look really did make an unbiased and informed prediction of Iranian behavior, it is easy to understand why Mattis is troubled. But if the exercise had to manufacture that Iranian response in order to achieve other exercise goals, it is less easy to understand his anxiety. In any case, he and his staff should consider why their assumptions -- which seem to require irrational Iranian behavior -- differ from Israeli assumptions.