At the end of Rosa Brooks' column on the demographics of the military today, she drops in this little tidbit:

Duke University researchers Peter Feaver and Christopher Gelpi analyzed data from the period between 1816 and 1992 and found that "as the percentage of veterans serving in the executive branch and the legislature increases, the probability that the United States will initiate militarized disputes declines by nearly 90%."

On one level, this makes intuitive sense—veterans seem less likely to romanticize war and all that. But I don't really know how Feaver and Gelpi arrived at those numbers, or how significant this result really is. Also, looking at the book's first chapter, it seems Brooks left out the next sentence:

At the same time, however, once a [militarized interstate dispute] has been initiated, the higher the proportion of veterans in the government, the greater the level of force the United States will use in the dispute.

Again, I don't know how much stock to put into this—the level of force used in a military conflict would seem to depend on many more factors than the makeup of Congress, but like I said, it has some surface plausibility. Meanwhile, Ben Adler argues today that liberals ought to shy away from valorizing military critics of the Bush administration. William Arkin said the same thing a few days ago, so did I back in April, and it still sounds right to me. Making a fetish of military officers and soldiers is a dangerous road to travel down—even though a 90 percent decline in "militarized disputes" would be nice (then again, it seems doubtful that veterans in Congress would vote to reduce the bloated defense budgets and sprawling network of overseas bases that help motivate so many military conflicts in the first place).