Pete Hegseth & the Firing of Generals

Since January 2025, Pete Hegseth has pushed out at least twenty-six generals and admirals and blocked or withheld promotions for about fifty-five more. Together that's more than eighty, roughly ten percent of the military's most senior officers. In September he announced a further ten percent cut to general and admiral billets, a twenty percent reduction in all, and he calls it accountability rather than a purge^[1]. The names are heavy: Joint Chiefs Chairman CQ Brown and Navy chief Lisa Franchetti in February 2025, Army chief Randy George, and on October 28, 2026, Lieutenant General Brett Sylvia, the J5, with no replacement named.

The case for the cuts is real. Supporters say the post-9/11 era bloated the flag officer corps and that trimming it restores civilian control; critics call it politicization. The honest read is probably both. The restructuring argument is defensible on its own; you can genuinely think the flag officer corps got too big. But almost none of the firings come with a stated reason, and you can't defend both the twenty percent cut and the selective enforcement without admitting the real filter is loyalty^[2]. Loyalty filters have a cost that shows up later, usually in a war you didn't plan for properly.

The J5 is the Joint Staff directorate for strategy, plans and policy, where political objectives become actual war plans^[3]. Sylvia was removed amid an investigation into alleged leaks about the Iran and Russia-Ukraine wars, which the Pentagon has referred to the Justice Department, and he's the second person pushed out of that chair in a year. That matters more than any single firing. It's happening during an active war with Iran, and once the midterms pass, the question isn't whether the military can fight; it's whether anyone left in the room has the authority and the experience to say no to a bad plan. Hegseth would say the chair can sit empty until someone clean walks in. But a plan written by nobody is a plan written by whoever shouts loudest in the room^[4].

Republican lawmakers aren't acting mostly because of incentives. Hegseth polls fine with Republican voters, a senator who holds hearings risks a primary from someone who'll say he sided with the woke Pentagon, and Senate Majority Leader John Thune isn't going to schedule anything that embarrasses the president days before the November 3 midterms. The pushback has come from outside that room: five former defense secretaries, Jim Mattis among them, called the first firings reckless in February 2025 and asked Congress for hearings, and Senator Jack Reed has said removing officers from a promotion list would be illegal if the reports were accurate. The cost of speaking is immediate and personal; the cost of silence is abstract and deferred^[5]. But silence isn't neutral: every week without hearings makes the precedent stickier.

The replacements say as much as the firings. Dan Caine, a career fighter pilot and a retired three-star, became Joint Chiefs chairman after Brown. The law asks a chairman to have served as vice chairman, a service chief or a combatant commander; Caine had held none of those jobs, so the requirement was waived. Christopher LaNeve, who commanded the 82nd Airborne and was Hegseth's own military aide, is acting Army chief after George, and he reportedly doesn't have the Senate votes to be confirmed. Other chairs, Sylvia's among them, just sit empty. The pattern is less 'new person, new vision' and more 'remove the old, leave the chair cold^[6].'

Nothing in recent decades matches it. Truman fired MacArthur in 1951, but that was one man over Korea policy. The bigger precedent is older: in 1941, as it got ready for war, the Army pushed out 195 Regular Army officers in about six months, most of them colonels and lieutenant colonels rather than generals. What's new now is the combination: mass removal, promotion blocks and empty key posts all at once, during a live conflict, with no stated doctrine guiding who stays. The closest analog might be Stalin's purges of the Red Army in the late thirties. Different motives, obviously, but the structural risk is the same: you don't find out what you lost until you need it^[7].