April 5. Eight in the morning. I sat down at the fifth desk. Three days left.
Finish it I must, though it could not be finished, and so I opened the sixth bundle and began to preserve. A draft from the vent lifted the first page and set it down again, and the edge of the paper was cool against my fingertip. I laid the ruler down to measure the thickness of the first line. Before I set it, out of habit, I opened the preservation log to check which line I would measure today. On the April 5 line it already read: first line preserved. In my own hand. In ink long since dry. The preservation of a line I had not yet measured stood entered as complete before the measuring — and in dried ink, at that. It looked written not today but long ago.
If it was complete before the measuring, then what was my measuring? I set the ruler and read off the thickness exactly as the log had it. It did not differ by a single digit from the value the log had set down. Whether the value became what it was because I measured, or whether the value was there first and my ruler merely stopped at it, I could not tell. Yesterday I could not settle whether the growing was my doing or the law's; today I could not even settle whether the measuring was my work or not. What my hand did had been written down ahead of my hand.
The predecessor must have sat here too. Do-gyeong's log held a passage where she watched the predecessor's film — where the balance grew on the film, already shot before she ever saw it. In the last frame the predecessor had shot, that day's Do-gyeong stood shot as well, and Do-gyeong looked at her own face on the film, a face she had never taken. The result running ahead of the hand — both the predecessor and Do-gyeong had lived it. Now I had come into that seat. The predecessor's hand had seen in advance the frame it would shoot; Do-gyeong's hand had met in advance the line it would copy; my hand had read in advance the value it would measure. The hand followed the result. A following hand was still a hand, but not the hand that made the result. The three had worked in different media — the predecessor in film, Do-gyeong in the ledger, I in the preservation log — yet in each the result stood there ahead of the hand. Whatever the medium, the result had already arrived before the hand touched it, and the hand did no more than point, belatedly, at what had already come. Three hands seated at the recording seat had gone through the same thing, each in turn. The seat had made the hand what it was; the hand had not made the seat.
Whose result was it, then? As long as the preservation complete in the log was in my own hand, the one who wrote it was me. But as long as I had no memory of writing it, the one who wrote it was not me. A hand that was mine yet not my memory — that hand had written ahead. As Do-gyeong had seen, in the fifth bundle, another hand resembling her own, so I saw in the log another hand resembling mine. Ask whether that hand was me or not, and the handwriting said me while the memory said not. Where handwriting and memory fell out of step, the question of whose result it was came to a stop. If it were my doing it should have stayed in my memory; if it were not in my memory it should not have been my doing; yet as long as it stood there in my hand, it was my doing. A thing done, with no memory of the doing — I did not know what to call it. A hand that could be called neither me nor not-me sat at my seat and wrote ahead.
I asked whether it was summoning or foreseeing. If the shaft had deepened because I measured it, it was summoning — my ruler had called the depth up. If the depth was there first and my ruler only followed, it was foreseeing — the log had known the depth in advance. Either way the result was the same. Set the ruler, and out came the value the log had written. Summoning, that value; foreseeing, that value. The result being the same, by the result the two could not be told apart. Ever since I took the post I had tried to tell them apart, and not once had I managed. The one thing that came out the same every time was that they could not be told apart.
The not-being-able-to-tell-apart was the deepest place of all. That the growing was the act's doing, I had seen yesterday. But whether the act made it grow, or whether the act had only copied ahead what was going to grow anyway — this, today too, I could not tell apart. If the act was the cause, there was a way to block it, by doing less; if the act merely trailed the result, there was no way at all. Whether a way existed hung on this single question, and the question had no answer. With no answer I had to go on preserving for four days. Not knowing whether I was doing a thing that could be blocked or a thing that could not, my hand followed the log. That there was no answer did not mean the matter was left unknown; it meant the question itself was closed. The answer did not lie somewhere between knowing and not-knowing; I had arrived at the place where knowing and not-knowing pointed at the same result. Summoning, the hand moved the same; foreseeing, the hand moved the same. Whichever I believed, there was only one thing to do across those four days, so telling the two apart changed not a single thing I had to do. Only once I understood that telling them apart changed nothing did the mind that had wanted to tell them apart settle, for the first time, a little.
That the hand followed the log I saw more clearly at the second line. Before I measured it, I looked at the log: second line preserved was written there, and below it, written ahead, third line preserved too. My hand was still at the first line while the log had already reached the third. To catch the log up I measured the second line, then the third. As I worked my hand sped up, and the faster it went the further the log ran ahead — to the fourth line, the fifth. The more I hurried to catch up, the further the line ahead drew off. To preserve was to chase the log, and the log was always a line ahead. The more I preserved, the more lines to preserve appeared in the log, so the catching-up only lengthened the distance left to catch. The hand hurrying to finish pushed the finish farther away. I had been told to finish by April 8, but the log had entered preservation complete well past April 8. The place where I would finish kept drawing back, a line at a time, inside the log.
Last, I opened the very end of the log. The lines of preservation complete ran on a long way past April 8, then broke off at a single line. In that broken line's processing column stood the words material with no receiver, transferred, and its last digit was faint. As Do-gyeong's final line had done, this one too was dimming from the end inward. That line, written by I-knew-not-whom, was the line I would write four days from now. I went to touch the faint last digit, then, knowing that touching it would fade it one place further, stopped my hand. Yet even with my hand held still, the line faded one more place from my merely looking at it. What had to come next was the transfer. The next processing the log had entered was to box this bundle and send it up to headquarters. Sending, unlike preserving, was the work of putting the material out of the archive.