Robert Kaplan has never shied away from bad ideas. A seasoned and sometimes shrewd observer of international affairs, Kaplan’s chief failing has always been his unwillingness to analytically retreat when he’s out of his depth—a weakness that often leaves readers stranded between mind-numbing banality and outright erroneousness.

Case in point: Kaplan’s new essay at Foreign Policy. Posing a reasonably interesting question—“Why is it so hard for strongmen to say goodbye?”—Kaplan offers an answer that is as intellectually flimsy as it is poorly presented. The reason, Kaplan argues, that Laurent Gbagbo, Muammar al-Qaddafi and Ali Abdullah Saleh just can’t bring themselves to leave political office is because…they’re “tribal warriors”!

The concept of warrior politics is familiar ground for Kaplan, who devoted an entire, and entirely absurd, book to the subject. Indeed, its only notable feature was the famous conclusion that “The short, limited wars and rescue operations with which we shall be engaged will…feature warriors on one side, motivated by grievance and rapine, and an aristocracy of statesmen, military officers, and technocrats on the other, motivated, one hopes, by ancient virtue,” a statement that stands out for being both nonsensical and patently wrong no matter how you slice it.

You might think that the book’s poor critical reception would make Kaplan think twice before resurrecting the “warrior” leitmotif in attempting to explore the Yemen, Libya and Cote d’Ivoire crises. After all, the notion of warrior politics, and attendant claims of ancient hatreds and the like, have been scoffed at and dismissed as being racist, unhelpful, and politically dangerous since at least the end of the Cold War.

But then you’d be wrong.

Things get off to a rotten start, and quickly. “By any rational standard,” Kaplan opens, “it would seem that the fighting and power struggles in the Ivory Coast, Libya, and Yemen should have been over weeks ago.” Really? What rational standard is that? And what precedent do we have to base it upon? Kaplan doesn’t bother with these sorts of considerations, but steams ahead to the observation that

the fact that they have already gone on as long as they have is an indication that there is a basic truth that those in the West fail to grasp about the individuals involved…[based on] reasoning [that] assumes that what divides these strongmen from their adversaries are issues as benign and susceptible to compromise as, say, Medicare and tax rates.

It’s not clear that anyone is assuming any such thing, but the basic point is fair enough. What, then, drives leaders? “They have been fighting for something far more age-old, basic, and less susceptible to compromise: territory and honor.” One need not bother pointing out Kaplan’s “the-barbarians-are-at-the-gates” racism to appreciate the fact that his driving thesis—that “their world is not one of institutions and bureaucracies [but] of dominating scraps of ground through dependence on relatives and tribal and regional alliances”—is already coming apart at the seams.

First off, according to Kaplan’s frame, “in such a world, figures like…Hosni Mubarak, are without virtue. They ruled in the Western style through institutions and bureaucracies, and when those institutions—the military and the internal security services—refused to shoot people in the streets, [they] had no choice but to meekly resign and quickly go into…exile.” Funny, I don’t remember Mubarak’s fall being quite so speedy. But this is largely beside the point. The real question here is: what does this have to do with anything? Nothing, it would seem, especially as Kaplan conveniently ignores the host of other cases where virtueless authoritarians operating through institutions and bureaucracies have stood fast in the face of popular protest—Iran, Bahrain, and Syria to name but three recent examples.

But it gets worse from there. According to Kaplan’s taxonomy of warrior thugs, “a figure like Gbagbo is especially despicable.”

In his mind, he fought an election and garnered close to half the votes. And those votes were not because of his position on this or that social or economic issue, but because of what he represented tribally and regionally…In places without sufficient economic development, like the Ivory Coast, elections often end up reifying differences based on blood and belief. To fight it out until he was cornered in the basement of his palace…is not a sign of moral weakness from his point of view, but of manly virtue.

Kaplan offers exactly zero evidence to support this claim, assuming that its truth is apparent on its face. Instead, he follows with the observation that

The same, of course, might be said of the sons of Saddam Hussein, Uday and Qusay, who were killed in a gunfight with US troops near Mosul in 2003—except that they, the spoiled-brat, gangsterish sons of the Stalinesque ruler, were by no means self-made men. Thus, they belong in a lower category of specimen than Gbagbo, Saleh, and Qaddafi.

Here again, Kaplan succeeds more in revealing his own class antagonisms and biased assumptions than he does in offering a coherent argument to explain the behavior of thuggish political elites under threat.

Seemingly sensing that readers might be scratching their heads in confusion, Kaplan gently admonishes his audience. “Remember, we are not talking about politicians so much as about warriors.” Oh, of course! How silly to forget! Except this is exactly what Kaplan does himself in the paragraph immediately following.

Take Saleh. The Western media labels the Yemeni president a recalcitrant tyrant whose stubbornness in clinging to power has, like Gbabgo in the Ivory Coast, threatened to unravel his country. [As if Yemen was the model of state stability before the recent protests.]…Saleh is clearly a man of steely nerves and subtle skill who, for decades, has dealt with levels of stress that would psychologically immobilize the most hardened Washington politico. The game he is playing now—negotiating the terms of his departure—is not just about him, but about the fate of his near and somewhat distant relatives. So, in a sense, who can begrudge him if he hangs on still longer, grasping for better and better terms?

Hold on. A moment ago, Kaplan was arguing that the manly ethic of tribal virtue militated against compromised solutions to political crisis. But now, Kaplan would have us believe that Salah is simply a crafty politician looking to work the angels for an optimum bargain. But never mind. Kaplan wraps up his discussion of Saleh by warning that “A few years from now, we may even look back on his rule as one of relative stability and cooperation with the West. Just because he deserves our condemnation now does not mean from an analytical perspective that he should be sold short.” Huh?

As for Qaddafi, “the fact that he has not gone quietly is a sign that he, too, is not fighting about any particular issues, per se, but about a vision of honor that strikes us as primitive, connected as it is to region, tribe, and territory.” I don’t know about anyone else, but Qaddafi doesn’t seem to me so much primitive as just plain nuts. Kaplan, however, isn’t all that interested in actually grappling with Qaddafi’s nature. Instead, he shifts gears entirely to set up new arguments of even greater incoherence.

And while we are on the subject of tribe and territory, it is important to recognize that the particular kind of tribalism that is one background factor in the rules of Qaddafi, Saleh, and Gbagbo is actually not a primitive, before-the-modern-state tribalism at all, but, as the late European anthropologist Ernest Gellner defined it, a tribalism that constitutes a conscious rejection of a particular government in favor of a wider culture and ethic…life under these men was hell, no doubt, but there was an identifiable logic to their madness, however much I have simplified it. Indeed, nobody captures the attraction of life outside the state as brilliantly as Yale University anthropologist James C. Scott in his book The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. Tribes today, Scott suggests, do not live outside history, but have “as much history as they require” in order to deliberately practice “state avoidance.” That is to say, tribes are rich in traditions and consequently do not seek the intrusion of government officialdom.”

This may offer an explanation of Qaddafi’s historic troubles getting control over the eastern regions of Libya, but hardly explains his own decision-making behavior. After all, for all intents and purposes, Qaddafi is the state, not an actor trying to escape it.

But no matter. Just when it seems like Kaplan’s analysis is about to crash and burn, he ejects from the cockpit and parachutes to relative safety with the limp and, at least in the case of Gbagbo, inaccurate conclusion that the three warrior rulers “have lived within this complex and ambiguous reality their whole lives and have thus not been state builders, yet another reason, in addition to the moral ones, that they have not found sympathy in the West. But that is no argument against trying to understand them.” That may be, but this essay surely offers good reason to give up trying to understand Robert Kaplan.