Chapter 6 The Plate From Tomorrow

The archive had a habit that Viveka had noticed her first week and had not thought about again until the night she went looking for a plate that did not exist yet. Every plate in the collection was sleeved, and every sleeve carried a date written in the hand of whoever had logged it, and the modern survey was filed in a cabinet of drives with a printed index that ran forward, night by night, because the schedule was planned months in advance. So the index always contained the future. If you opened the index at the current date and read forward, you found the next night's fields and the next week's and the next month's, each one a plan not yet executed, each one written in a calm and ordinary type. She was reading that forward edge on the sixth night, looking for the next time field seven came up, when she saw that the entry for the following night had already been stamped with a plate number. A plate number is assigned when a plate is made. There is no reason to assign one to a night that has not happened. And yet there it was, in the index, the number and the field and a small note in the margin: *good seeing.* She went to the cabinet. The plate was there. She pulled it and took it to the light table and turned on the lamp, and there in the cold glass were the stars of field seven, sharp and clean, exactly the field she would photograph the following night, taken, according to its own sleeve, on a date that was still eleven hours away. She checked the sleeve. She checked the index against it. She checked the telescope's log of exposures, and there was no record of any exposure on that night, because the night had not come. But there was the plate, and there was the field, and in the small corner where the object lived, the object was there too, and it was not quite where it had been. It had moved. For the first time since she had begun measuring it, RV-417 had a position that differed from the one in the catalogue, shifted a little to the east by an amount she could measure: eleven milliarcseconds. She sat with that number for a long while. The object had been perfectly fixed for thirty-five years of plates. Now it had moved, and it had moved at exactly the rate the survey's own model would have predicted if the object were an ordinary distant star with an ordinary proper motion, as though the thing had finally decided to obey the rules, and had chosen to begin obeying them one night early. She checked the cabinet log next, the printed sheet on which every removal and every return was initialled, and found the plate logged out once, that morning, by her own hand, and not at all before. Then she checked the index again, and the index still carried the plate number against the following night, and she went round the loop a fourth time and found no door in it. A thing that has not happened cannot leave a record. That is the whole of routine, and the whole of trust, and the whole of an observatory. She had spent twenty years working inside that rule, checking it and relying on it and never once needing to prove it, and here, in a cold glass plate on a warm table, sat a record of a night that had not yet been spent, made by instruments that were still waiting to be pointed. She considered, seriously and in the end without belief, the possibility that she was tired. She considered it because it was the explanation the mountain would give her if she asked, and because it had the advantage of being the one explanation she could test. She did not know what to do with the plate. She photographed it with her own camera, because she had learned to trust her eyes only when they left a copy, and then she put it back in its sleeve and put the sleeve back in the cabinet and closed the drawer. The trouble with a plate from tomorrow is not that it is impossible. The trouble is that it is ordinary. It is sleeved and numbered and filed and initialled like every other plate, and it sits in its drawer with the others, and if she had not happened to read the index forward on the one night she read it forward, she would never have known, and the season would have closed with the anomaly recorded nowhere and the survey one clean epoch longer. She kept returning to that, because it was the part that had a purpose in it. Whatever had made the plate had not made it to frighten her. It had made it to be filed. The next night she took the scheduled exposure of field seven. The seeing was good. When the reduction came up, the object sat exactly where the pre-dated plate had put it, eleven milliarcseconds east of the catalogue, as though the plate had not predicted the observation but instructed it. For the rest of that night she did something she had never done in twenty years of observing, which was to check her own senses. She took a standard star and measured it four separate ways and wrote the four results down before she compared them. She had Lisbet measure the same star blind, from a frame she chose. She measured a field she had never looked at before and could not check against memory. Every result agreed with every other, and the agreement should have reassured her, and did not, because the one thing she could not test was the instrument that had made all four measurements, and that instrument was her. She took that to Lisbet in the morning and watched the junior observer go very still. "You pulled a plate for a night that hadn't happened." "Yes." "And it was there." "And it was there, and it was right, and it was right the way a plan is right." Viveka put the two measurements side by side on the console: the plate's position and the night's observation. They agreed to the last digit. "A prediction and an observation should agree within the error. These agree exactly. Nothing agrees exactly." "So the plate was made after the night and dated wrong." "Six ways that would have to happen, Lisbet. The sleeve, the index, the cabinet log, the notebook entry, the schedule stamp, my own reading of it. All wrong in the same direction by the same amount." She shook her head. "This station does not make that many mistakes at once. This station makes one mistake and files it." Lisbet looked at the console for a while. Then she said the thing that Viveka had been avoiding, because it was the only thing left. "Then the plate wasn't made late. It was made early. Something made it before the night, and the night did what the plate said." Neither of them said anything else. In the galley down the corridor, Kolstad was doing the dishes, and the water ran, and someone laughed at something, and the ordinary sounds of a small station went on being ordinary in the way that had begun to feel like a performance. Viveka took the plate's number out of the index that afternoon, with a pencil, and wrote the number in her notebook instead, and then she went to the cabinet and moved the plate to a different drawer, out of order, where no one would look for it. She could not have said why. She knew only that she had begun to keep two records of the sky, one that the station could read and one that it could not, and that she had done it without deciding to, the way you close a door behind you in a house you do not entirely trust. ---