The Useful Fool: Why Power Has Always Preferred Devotion Over Brilliance
In the spring of 1812, Napoleon Bonaparte dismissed his most capable foreign minister and retained advisors of considerably narrower gifts. The decision puzzled contemporaries and has puzzled historians since. It should not puzzle anyone who has spent time with the historical record. The logic was not strategic in the conventional sense. It was existential. A subordinate who might be right when you are wrong is a subordinate who might one day be right instead of you. That distinction, across five thousand years of documented power, has proven fatal to more brilliant careers than incompetence ever managed.
The inner circle is not selected for its capacity to make the leader effective. It is selected, first and always, for its capacity to make the leader safe.
The Economics of Loyalty
To understand why leaders consistently elevate the devoted over the capable, it helps to think in terms of risk rather than merit. From the perspective of someone who holds power, subordinates present two categories of threat: the threat of failure and the threat of displacement.
An incompetent loyalist produces the first threat. Bad decisions get made. Opportunities are missed. Occasionally, disasters occur that could have been avoided with better counsel. These are real costs. They are also, from the leader's perspective, manageable costs. Failures attributable to a loyal subordinate can be absorbed, blamed on circumstance, or blamed on the subordinate himself—who, being loyal, will accept the blame and remain. The leader survives the failure.
A capable skeptic produces the second threat. And the second threat is categorically different. A subordinate who is more competent than you, and who has demonstrated a willingness to say so, is a subordinate who has already shown other stakeholders—rivals, factions, foreign powers, the public—that the leader's judgment is not uniquely reliable. That demonstration, once made, cannot be unmade. It creates an alternative center of gravity. And alternative centers of gravity, in the history of political power, have a tendency to become alternative centers of power.
The loyal incompetent cannot do this. He is dependent. His elevation is your gift. His survival requires your survival. His interests and yours are structurally aligned in the way that matters most: neither of you can afford the other's fall.
From the Pharaoh's Court to the Cabinet Room
This dynamic appears in the earliest administrative records we possess. Egyptian papyri from the New Kingdom period document the elevation of relatives and personal retainers to positions of significant administrative authority—positions for which, the same records sometimes candidly note, they were not the most qualified candidates available. The logic was not concealed. It was, in the framework of ancient Egyptian governance, straightforwardly rational: a man whose position depended entirely on the Pharaoh's favor had interests that were structurally inseparable from the Pharaoh's continued authority.
The Byzantine court systematized this instinct into institutional practice. The use of eunuchs in high administrative roles was not primarily about trust in the abstract sense—it was about structural dependency. A eunuch could not found a dynasty. He could not, in the most literal sense, threaten succession. His entire social existence was a function of his patron's existence. This made him, from the emperor's perspective, a nearly ideal subordinate: capable enough to be useful, structurally incapable of being dangerous in the way that mattered most.
American political history offers less dramatic but functionally identical examples. Presidential administrations across the twentieth century show a consistent pattern: the advisors who survive longest are rarely those whose counsel was most accurate. They are those whose counsel was delivered in a manner that did not require the president to publicly acknowledge error. The distinction sounds subtle. Its practical consequences have been enormous.
When the Brilliant Subordinate Speaks
George C. Marshall, by most accounts one of the most capable military administrators in American history, survived in positions of authority longer than many of his contemporaries in part because he understood the political economy of advice. He delivered uncomfortable assessments privately, framed his disagreements in terms of the superior's goals rather than his own judgment, and declined to build an independent public profile that would have made him a rival rather than a resource.
This is not a story about Marshall's humility. It is a story about Marshall's sophistication. He understood something that many capable subordinates do not: the value of your advice to a leader is not simply a function of its accuracy. It is a function of whether the leader can accept it without feeling that accepting it diminishes him. Advice that is correct but threatening tends to be rejected. Advice that is slightly less correct but unthreatening tends to be adopted. Over time, the advisor who understands this survives. The advisor who does not understand it is replaced—often by someone less capable, and often at significant institutional cost.
The leader who made that replacement was not making an error in the conventional sense. He was accurately calculating his own survival interests and acting on them.
The Institutional Cost
The tragedy embedded in this pattern is that it is individually rational and collectively ruinous. A leader who systematically selects for loyalty over competence is making a reasonable decision about personal survival and an increasingly unreasonable decision about institutional effectiveness. Over time, the inner circle becomes progressively less capable of delivering accurate assessments of reality. The leader becomes progressively more insulated from information that would complicate his preferred conclusions. The institution—the army, the administration, the corporation—begins to operate on the basis of what the leader wishes were true rather than what is.
Rome's late imperial period offers the clearest extended case study. The progressive elevation of loyalists over capable administrators did not destroy the empire in a single moment. It degraded the empire's capacity to respond to genuine threats over the course of generations, until the gap between institutional reality and institutional self-perception became unbridgeable.
Modern American political institutions are not immune to this dynamic. They are subject to it in precisely the way that all human institutions have always been subject to it: through the accumulated individual decisions of people who correctly understood that their survival required surrounding themselves with those who would not threaten them.
The Record's Verdict
The historical record does not suggest that leaders who prioritize loyalty over competence are foolish. It suggests that they are rational actors operating within a system of incentives that consistently rewards personal survival over institutional performance.
What the record also suggests is that institutions built on this foundation are fragile in proportion to how thoroughly they have implemented it. The loyalty premium is real. So is its interest rate. And across five thousand years of documented governance, the compound cost of that interest has brought down more regimes than any external enemy managed to.