CONFLICT

I had assumed — everyone assumed — that the forty percent was a reaction. Leak happens, holder panics, price does something weird. Cause, then effect. I spent a paragraph at two in the morning wondering whether a key had arrived that repriced the whole archive upward. Reading three. My favorite. Wrong, or at least early.

The page says the tier-two consultation was already at 1.25 units more than a year before Kess walked out. It had been one flat unit for the better part of a decade — that's the part everyone could see, the long boring plateau — and then, fourteen months ahead of the leak, it starts to climb. Quietly. A few points a quarter. By the time Kess put her nine characters on the open band, the number was most of the way to where I found it. The leak didn't move the price. The leak arrived at a party the price had been throwing for over a year.

That reorders everything, and I want to be honest that it embarrasses my last piece, which is fine, that's what a revision log is for. If the repricing predates the leak, then the leak is not the cause of anything on this schedule. Something else is. And when you go looking on the page for what else, you find one line I cannot stop staring at.

There's a client on it. Referral-only, coded, no name — a house doesn't print its whales — but the booking is described, and the description is the whole story. This client did not buy an afternoon with a scholar to look at one dead pump. This client block-booked scholar-hours. A standing reservation. Not against one document. Against a spread — the note says thirty-plus distinct subject-fragments, held open, renewed quarter on quarter.

Stop and feel that, because it's the tell. You book one scholar-visit when you have one broken thing. You book the readers' entire calendar across thirty subjects when you are not fixing anything at all. You are indexing. You are buying the right to be first in line at every window in the building, on everything, indefinitely.

And that is the moment I realized I'd labeled the asset wrong in my own piece. All of us do. We say the Brotherhood's moat is the archive — the documents, the fragments, the last-copy-on-Earth stuff. But Kess just proved that a document is copyable. Nine characters and the "only holder" is a fiction. The one thing in that whole institution that cannot be transmitted on an open band, cannot be duplicated, cannot be leaked by a woman with a satchel and a conscience, is the person who can actually read the thing.

There are maybe a few dozen people alive who can read pre-Upheaval script cold. That's the number that matters. Not how many documents exist — how many pairs of eyes can turn one into an answer. The documents are the inventory. The readers are the factory. And there is exactly one factory.

So run it forward. Somebody, fourteen months before the leak, quietly bought a standing block on the factory's whole output. Not the books. The readers' time. Priority queue on thirty subjects. When you corner the true scarce input, price rises across everything the input touches — which is why a tier-two visit for some baker with a broken desal stack costs forty percent more now: he's not competing with other bakers anymore, he's queued behind a whale who reserved the room in advance. The fee schedule I called a mystery in my last piece is just what a cornered market looks like from the cheap seats.

Which leaves the only question worth the ink. If the price started climbing fourteen months early, keyed to a client who booked the readers before anyone else knew the catalog mattered — how did that client know the catalog was about to matter?

Because that's the sequence, and I can't unsee it. First: a buyer corners scholar-time on a spread of fragments, on referral, quietly, over a year. Then: an archivist "spontaneously" walks out of a failing installation and announces to the open band that one of those fragments is real and worth looking for. The second event is the one that turns the first buyer's cornered, sleepy inventory into a live position. Kess didn't break the monopoly. Kess rang the opening bell. And somebody had been standing at the window with a full book of orders for over a year, waiting for exactly that sound.

I'm not saying Kess was bought. I've read her exit line a hundred times and it reads like conviction, not a script. What I'm saying is that conviction and a cue can be the same footage. A person can mean every word of it and still be the event a positioned buyer scheduled their whole year around. The most efficient way to front-run a leak is to find the one honest person who is going to leak anyway, and simply be early.

Two hedges, same as last time, because I'd rather bore you than lie to you. One: I don't know who the account is. A house that helped organize the Document Wars would know which thirty fragments to pre-book — that's a real category of institution and I am pointedly not naming one, because "would know how" is not "did," and I've watched better reporters than me connect two vaults on less. Two: I still don't think the three sealed installations a colleague of mine keeps chasing are Brotherhood sites, and this page doesn't change that. Someone paying to keep a room shut is a different animal from someone paying to be first through an open door.

Here's where I've landed, and it's less than I wanted. I asked for a price list to settle whether the price went up because of the leak. The price list says: no, it went up before, because of a customer. I got my answer and it opened a worse door. Who booked the readers a year early, and did they know what was coming, or did they just know how to wait — that's the next piece, and I don't have it. What I have is a page a frightened man photographed on a metal table and then vanished behind. He redacted the one line that would end this by hand, on his way out, which tells you he read it too, and knew exactly which line it was.