The third rumor was that Rowe had taken kickbacks from a supplier, specifically from the linen company that served the hotel. The reviewers examined the linen contract, and the invoices, and the payments, and they found nothing: the prices were in line with the market, the deliveries matched the invoices, and the supplier's owner, when contacted, said he had never paid Rowe anything, and that he barely knew him. The reviewers marked the rumor unsupported, and they noted, in the report, that the service-charge diversion had not been a vendor scheme at all, and that it was important to say so, because a report that accused a man of the wrong kind of wrongdoing would let him answer the wrong charge and walk away from the right one.
The fourth rumor was that Rowe had pocketed guest deposits for banquet cancellations. The reviewers examined the banquet deposit records for two years, and they found that deposits had been handled correctly, refunded on time, and accounted for in full, and that the rumor had apparently grown from a single incident in which a deposit had been delayed by a bank error, and the delay had been explained, at the time, to the guest, who had accepted the explanation. The reviewers marked the rumor unsupported, and Ramona, when she read the section of the draft report that dealt with rumors, felt a small, stubborn satisfaction, because a review that clears a person of what he did not do is a review that can be believed about what he did do, and because fairness, in the end, is not a courtesy extended to the powerful. It is a discipline that protects the powerless, by making sure that when the findings come, they cannot be dismissed as exaggeration.
There were other rumors, smaller ones, about favoritism and about who had been promoted and why, and the reviewers checked those too, and some of them turned out to be true in a limited way, and some turned out to be misunderstandings, and all of them were recorded, with their status marked, in the working papers. Ramona insisted on this, even though it made the report longer, because she had learned, over eleven hotels, that the fastest way to lose a staff's trust was to tell them that a thing they believed was false without showing them how you had checked, and that the second fastest way was to tell them that a thing they believed was true without checking it at all. The report, when it came, would be two hundred and ten pages long, and a part of its length would be this: the careful separation of what was proved, what was not proved, and what was proved to be untrue. She thought that was the right length for a document that would decide people's livelihoods, and she thought that anyone who wanted a shorter report had not understood what a report was for.
The report took five weeks to write after the mandate, and it was two hundred and ten pages long, and Ramona wrote it the way she wrote everything: slowly, with the records open beside her, and with a rule that no sentence would go into the report unless she could point to the document or the interview that supported it. Her two reviewers drafted sections, and she rewrote them, and they argued, in the conference room, about wording, about whether a particular schedule cut could fairly be called retaliation, about whether a particular complaint had been diverted or merely handled informally, and about whether the service-charge finding should be stated as a diversion or as a misapplication, and they settled these arguments the way they settled everything: by going back to the record, and by asking what the record would support, and by choosing the weaker word when the stronger word could not be proved. The report, when it was finished, did not read like an accusation. It read like an inventory, and that was deliberate, because an inventory is harder to argue with than an accusation, and because the ownership group, when it read the report, would need to act on it, and action requires a foundation that does not shift when it is pressed.