Dominic Cummings (right) is Prime Minister Boris Johnson’s most influential adviser.Daniel Leal-Olivas / Getty

In 1989, the billionaire investor Warren Buffett wrote a letter reflecting on what was then his 25-year stewardship of Berkshire Hathaway. His “most surprising discovery” was “the overwhelming importance in business of an unseen force that we might call ‘the institutional imperative.’” Buffet told investors that he initially thought “decent, intelligent, and experienced managers would automatically make rational business decisions.” In fact, he discovered, this was untrue. The crucial thing was the system.

It was almost a law of business, Buffett wrote, that institutions will resist change, waste time, produce evidence to support the whims of whoever is in charge, and mindlessly imitate the behavior of rival companies. The lesson he took from his insight was to organize his company in ways that minimized the dangers of systemic failure and to invest in other companies that also seemed to understand this risk.

Buffett’s law has entered the annals of business theory, and been held up by an array of leaders—among them Dominic Cummings, Prime Minister Boris Johnson’s most influential adviser.

Cummings’s belief in this worldview goes a long way toward explaining the ongoing tension between the U.K. government and the institution it is asking to enact its agenda: the British Civil Service. The tension exploded into public view this weekend with the abrupt resignation of the most senior official in the Home Office following repeated disputes with his political master, Home Secretary Priti Patel, who has been tasked with introducing an entirely new immigration regime in just 10 months. It was also on display in the revelation that the British government opted to leave the European Union’s pandemic early-warning system, despite calls from health experts to remain. But while the clashes, particularly at the Home Office, appear to have been personal and specific, involving accusations of bullying and leaking, or of placing ideology over pragmatism, they are also systemic and general, revealing a clash of ideas that is only likely to intensify, and not just a clash of personalities.

At the heart of the row is the idea set out by Buffett in 1989, that it matters not the “venality or stupidity” of the government minister or civil servant in question (depending on your view of who is in the right), but the hardwiring of the government machine itself. In Cummings’s view, the Civil Service, like every other government bureaucracy, is unable to shake Buffett’s institutional imperative toward inaction, time-wasting, false certainty, and mindless imitation. The machine, Cummings believes, is unable to bring about the kind of change voters want—therefore, the only way to change the outcomes that voters dislike is to change the system. Regardless of who behaved badly in the dispute between Patel and her senior civil servant, this clash of ideas is the key underlying tension.

Cummings’s perspective is best understood not through a traditional left-wing or right-wing political prism, but philosophically. It is revolutionary in nature, seeking systemic change, not just policy reform, and means that individual bust-ups are best understood not on their specific merits alone, but as part of a wider battle over the nature of government itself.

To better understand this clash, it is instructive to return to Cummings’s writings on previous British governments that have found themselves frustrated by the small-c conservative nature of the Civil Service. Both Tony Blair and David Cameron, for instance, sought ways to more efficiently control the machine, to shake it to life to produce the outcomes that the government wanted. Blair chose to centralize and impose targets; Cameron criticized this approach from opposition, only to imitate it in power.

After a crisis following the outbreak of foot-and-mouth disease early in his first government, Blair discovered “the wonders of Cobra,” Cummings writes, referring to the U.K.’s disaster-response system, named after the chamber in which it is held—Cabinet Office Briefing Room A. The military response to the crisis showed Blair what could be achieved through command and control, and when further crises emerged, Blair turned to Cobra. Today, Cobra is the go-to response taken by prime ministers when a crisis occurs. Johnson himself was seen to take control of the growing concern over the new coronavirus only when he convened his first Cobra meeting on the matter on Monday.

For Cummings, though, the reliance on Cobra shows not a system working but “a symptom of Whitehall’s profound dysfunction.” The problem, in his view, is the very thing the Civil Service is most proud of: its apolitical permanence. The Civil Service’s defenders say this is what makes it the best in the world, producing “Rolls Royce” diplomats and officials who are experts in their field; free from short-term political pressures; able to give honest, impartial advice; rounded by experience across government. Cummings counters that this is exactly what is wrong with the Civil Service, entrenching its destructive imperative to preserve the status quo, unable to imagine an alternative to the way it works, wasteful of time and resources, and ultimately incapable of adapting to sudden change it never foresees—Brexit chief among them.

The point is not that the Civil Service should have forecast these events, but that it should confront its inability to predict the unpredictable. To Cummings, politics is like the weather: Forecasts are valuable, but limited. In politics, he wrote: “Unfathomable and unintended consequences dominate. Problems cascade. Complex systems are hard to understand, predict and control.” The key is the ability to adapt—and to adapt, one has to embrace the reality of uncertainty. Government machines instead are built to try to eliminate uncertainty. What is required is a more scientific approach to government, learning from experience as Buffett proposes. The British government should take its cue from the human immune system, Cummings argues, where there is no plan or central control; its strength lies in reacting to attack, experimenting, doing whatever works, discarding what doesn’t. Such a system is messy but adaptable, and therefore stronger.

Here we come back to Patel, the theory of government, and Brexit itself.

Government supporters argue that the civil servant who quit in protest was acting as a block to democratic change: a post-Brexit immigration system. He has responded that he and his staff were subject to “unreasonable and repeated demands” from the government, as well as personal abuse. Brexiteers therefore have held up Patel as the hero for taking on the system (#IStandWithPritiPatel was trending on Twitter this week); Remainers have held up the official as the hero for standing on the side of the experts warning that such demands were impossible. This argument, though, is meaningless if it is the system that is at fault, not individuals within it.

The row also reveals the deeper philosophy of Brexit, which drives this Johnson administration and has yet to be fully grasped by those who routinely show exasperation at its apparent refusal to listen to expert advice. Johnson, they say, is pursuing a future that makes no sense, one in which sovereignty is prioritized over economic alignment with the EU, meaning that Britain will be poorer than it needs to be.

But this misunderstands the core of the Johnson-Cummings project. It is not that they disagree with experts’ forecasts, or that they are attempting to be populist. They actively reject this model of government, believing it to be systemically and empirically flawed. They argue that Britain needs to free itself from centralized bureaucratic control, rather than rely on it, to be able to react both to domestic crises and the ever-changing international environment. This is a project to remake Britain into a country agile enough to adapt quickly to the dramatic change that is inevitable and unpredictable, not to perfect an existing system that avoids unwanted shocks.

This, then, is not a row about what immigration regime Britain should adopt after Brexit or how quickly it can be put into place, but about the very nature of government and how to be a successful country in the 21st century. It is an experiment. Sometimes experiments succeed and sometimes they don’t. The question, taking Cummings at his word, is whether Britain is institutionally capable of adapting to make it work.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.