Nine of twenty-two. She has the number. She will never have the names. Mio Tashiro came from Sendai to London to study econometrics, because she had believed since she was nineteen that a number could not be rude to you. What she wanted was a place where being right about a number was the point. What she gets is ten weeks on the ninth floor of a bank at Canary Wharf. An orange lanyard, so that colleagues will know to offer assistance. A desk in the middle of the floor with no drawers, because flexible seating cannot have drawers. And on the first Thursday, an invitation to a welcome dinner, sent to nine of the twenty-two summer analysts. Seven of the nine are women. Mio, who has spent two years being taught at some expense exactly how not to conclude things, works out the odds and writes them down and does nothing. Then she is handed eighteen months of building access data for a floorspace study — fourteen million rows recording where every person in the building was, to the nearest lift lobby, at every moment for a year and a half. She is given it because it is boring, and she is an econometrician, and nobody thinks about it for more than nine seconds. She does not go looking for anything. It comes out of the residuals, which is where things come out of. Return Offer is a novel about an institution that has never lied, never broken a rule, and never once been asked the right question. There is no villain in it. The desk head is courteous and remembers her name. The second-year analyst is warm with both hands. The Head of Culture believes her, within eighteen hours, and spends sixteen months rebuilding the framework so the next one will not have to be moved. Every door, every policy, every deadline and every dinner was installed by a competent person for a defensible reason, and most of them by a kind one. It is the assembly that sorts people. And an assembly has no defendant. Cold, exact and quietly furious — a literary novel about measurement, complicity, and the distance between knowing the size of a thing and being able to prove it.
