The Generals has one of the best qualities a general nonfiction book can have: it’s about a specific topic that it covers well, but its lessons and ideas also transcend its topic and apply to many others. Let me explain. Take this section, about General Patton:*

Even now, more than six decades after his death, Patton remains one of our most remarkable generals. ‘You have no balance at all,’ Marshall’s wife once scolded the young Patton, correctly, years before World War II. Maj. Gen. Ernest Harmon, one of his peers, wrote that he was ‘strange, brilliant, moody.’ The blustery Patton behaved in ways that would have gotten other officers relieved, but he was kept on because he was seen, accurately, as a man of unusual flaws and exceptional strengths. Marshall concluded that Patton was both a buffoon and a natural and skillful fighter.

Knowledge, skill, and expertise in one domain don’t necessarily transfer to other domains. A brilliant physicist may be a terrible marriage therapist, and vice-versa. Someone who is a “buffoon” might also have a compensating skill that makes up for their possible deficits. Paul Graham implicitly writes about this in Is It Worth Being Wise?:

‘wise’ means one has a high average outcome across all situations, and ‘smart’ means one does spectacularly well in a few. [. . .] The distinction is similar to the rule that one should judge talent at its best and character at its worst. Except you judge intelligence at its best, and wisdom by its average. That’s how the two are related: they’re the two different senses in which the same curve can be high.

A lot of people seem to have trade-offs between peaks and averages. Steve Jobs comes to mind: Walter Isaacson’s biography is rife with examples of Jobs being wrong, cruel, and occasionally outright stupid. His lows were low. But he got big, important stuff right—and not just right, but very, spectacularly right. He found (or made) the right environment for his skills. It’s almost impossible to imagine Jobs being a good employee at, say, Wal-Mart, or any large company that values homogeneity over creativity.

It’s obviously possible to have high averages and high peaks, but that doesn’t appear to be common. Really spectacular peaks often come in unusual packages. Those unusual packages are often easy to dismiss by someone not paying attention.

Unfortunately, as Ricks points out, America since the Korean War hasn’t judged its generals by their peaks or their averages: in fact, we haven’t judged generals on their competence much at all. That’s a tremendous, underappreciated problem. In Ricks’ description, the generals cut from the Marshall style were primarily “team players” who needed to work effectively with others and defer to the group. That’s not necessarily a bad thing; as Ricks says:

Perhaps those who rose highest in World War II were organization men. But for the most part they were members of a successful organization, with the failures among them weeded out instead of coddled and covered up. That would not be in the case in our subsequent wars, in which it would be more difficult to know what victory looked like or even whether it was achievable.

Different time periods reward different forms of industrial organization. If World War II rewarded “organization men,” many of today’s organizations reward people who figure out the weaknesses of large organizations, and then offer alternatives. But that can’t happen in the military, where the closest analogue to startups might be defense contractors and private, Blackwater-style armies. Those, however, have their own problems.

There’s also an analogy to teaching: almost no public school teacher is fired, ever, for bad teaching. Not being able to fire transparently terrible teachers is an impediment to getting better teachers, as almost anyone who’s ever been in a public school knows.

Organizations also need to focus on making sure that they’re focused on their major purpose, not on primarily serving the interests of the people inside them:

Trying to be fair to officers can be lethal to the soldiers they lead on the battlefield. The Army was using the Korean War to give the staff officers of the earlier war ‘their chance’ to command in combat—with disastrous results. Well before Chosin, the Army had recognized that it had a problem with inexperienced combat leadership in the war.

The problem is “inexperienced combat leadership,” but the solutions became worse in some respects than the problem itself. Fairness to one group can mean extreme unfairness to others, who often have much less of a voice. No one speaks for the enlisted men who are led by incompetent generals. (No one speaks for those led by an incompetent president, either, but that’s a separate issue related to larger American society.)

Misaligning incentives creates a deeper sense of rot; Ricks says that generals, by the post-Korean-War era,

were acting less like stewards of their profession, answerable to the public, and more like keepers of a closed guild, answerable mainly to each other. Becoming a general was now akin to winning a tenured professorship, liable to be removed not for professional failure but only for embarrassing one’s institution with moral lapses.

Notice what this says about Ricks’s view of the university: by comparing one system that advances mediocrity with tenure, he implies that tenure advances mediocrity. He doesn’t go on to explain why he uses the metaphor, because he assumes that his readers already believe as much. But tenured professors aren’t putting their students in life-or-death situations, and students can choose to pick a different department or university. Service members can’t. During World War II, as Ricks says, the road to victory and home led through Berlin and Tokyo. In recent wars, the road to victory has been murkier, the politico-military establishment mostly hasn’t selected generals adept at operating in the murk. The consequences are clear.

The Generals is too detailed for people who aren’t deeply interested in military affairs and history. It probably isn’t detailed enough for those who are immersed.

But it’s also the best intellectual explanation of why one should be wary of enlisting in today’s American military: you might get killed by someone incompetent but unaccountable on the basis of performance. Contemporary generals who lose wars and cost soldiers their lives are fêted. They “retire” to lucrative consulting gigs with defense contractors and lobbying firms. The soldiers are disabled or dead. To me that argues against becoming a soldier or junior officer. In most businesses, if you think your boss is an asshat, you can quit and start a rival firm. In the military, obeying is the only option, and no one is making sure that your boss is actually good at his job.

EDIT: B.J. Khalifah has an interesting letter in The Atlantic:

Thomas Ricks overlooked something important. Sadly, nobody becomes a general (or equivalent) in the military until they have served for many years. Most colonels are 50 by the time they get promoted. Many younger officers have experience and drive; as a group, they adapt well. Older officers are more cautious, members of the “cover your ass and do not make waves” category. They know how to manipulate the good-old-boy game. The service should be, but is not, a strict meritocracy. In effect, it follows union-style rules of seniority and time in grade. From second lieutenant to first lieutenant to captain is automatic. Some lousy officers have made it past captain to become major by being on court-martial or combat duty when they are promoted. The rules are not negotiable.

This contrasts hugely with startup and good corporate cultures, which judge people almost purely on merit. Successful startups have famously been founded by 18 year olds. Even law firm partners can be promoted within as little of five years of hiring, while associates frustrated by a firm’s practices can start their own. The military apparently doesn’t do that, and I haven’t seen any evidence that 50-year-old generals will necessarily be better than 26-year-old (hypothetical) generals. Certainly among startups this isn’t true.

The comparison isn’t perfect—markets reward innovators for making things people want, and the military doesn’t have a clear feedback loop. But at the moment almost no one is even discussing the issue, or making the comparison.

* The movie Patton is also remarkably good, especially the speech at the beginning. Patton doesn’t have the American character down correctly—Americans don’t love the sting of battle unless we’re provoked—but the speech demonstrates a lot about the man doing the speaking.

The bit about loving a winner and not tolerating a loser is also fascinating in light of The Generals: we’ve tolerated a lot of losers, like Donald Rumsfeld and Tommy Franks, and sacked winners like Eric Shinseki.