The Row That Doesn't Sum: CKM Unitarity's Deficit, Priced

The weak force keeps a three-by-three ledger of how quarks change into one another, and every row must sum to one. The first row is the best-kept row in the book. Its largest entry, |V_ud|, from fifteen superallowed nuclear beta decays, reads 0.97367, give or take 0.00032. Its second entry, |V_us|, forks by road: the semileptonic kaon road reads 0.22330, give or take 0.00053, while the leptonic road implies 0.22500, give or take 0.00050. Add the row up and it does not sum. With the semileptonic value the row totals 0.99791 — short by 0.00209, at 3.1 sigma. With the leptonic value it totals 0.99867 — short by 0.00133, at 2.0 sigma. With the Particle Data Group's scaled kaon average it totals 0.99836, short at 2.2 sigma, matching the Group's printed 0.9983 and its verdict of about two sigma. This paper prices the shortfall door by door: the universal radiative correction would have to fall 8.3 of its own error bars to close the row alone, and below even the older Marciano-Sirlin value it replaced; each kaon road would have to move four to nine of its own error bars; and new physics must enter at the per-mille level — a right-handed W coupling to up and strange quarks of about two parts in a thousand in effective-theory units, or an electron-neutrino mixing with a MeV-scale sterile partner of about one part in a thousand in the mixing squared. The claim is narrow: the deficit is a dated, priced tension in the present record, two to three sigma depending on the inputs spent — not a discovery claim. Grade: DEFICIT STANDS — PRICED, TWO TO THREE SIGMA, CHEAPEST DOOR A THEORY INPUT. This paper stands outside my work on the vacuum catastrophe; it draws nothing from that work and repairs nothing in it.