July 16. In the morning I opened the copy first.
Page seventy-six: July 16, forty-five. Yesterday's structure pushed along another day. Today's date with tomorrow's figure.
Below it page seventy-seven stood as it had appeared yesterday. The date column was empty. In the figure column stood not a figure but words. The figure of a day not counted.
I saw those words and went home. On the way and waking in the night I thought about them. It is the first time in six months the copy has put words in a figure column. That something which is not a figure has entered the column where figures go means either that what the column waits for is not a figure, or that a figure not yet settled has been called by a name.
The figure of a day not counted is a figure that requires a day not counted. For six months I have counted every day without missing one. On the ninth I held out two hours and counted at the end of them. There has been no day left wholly uncounted.
So the column is still nothing. If I count today it will be empty today as well; if I do not, something will enter it. To see what enters, I must not count.
Today I will not count.
Nine o'clock came. My hand moved first. It is the motion of nine o'clock for six months, and the hand had the order by heart. Putting it on the keys and taking it off again took a long while. Off the keys it did not know where to go and sat on the edge of the desk.
That not doing should take effort was strange. Counting takes ten minutes. I spent the hour from nine to ten not doing those ten minutes. Not doing took longer than doing.
At ten I handled eight requisitions. I wrote the replies and set four reels on the reader. The work finished early today. The ten minutes of counting and the thirty or forty minutes after it, opening the list one item at a time to confirm, were gone whole. With them gone I saw that they had held the middle of the day. For six months my day was divided front and back around the counting. Take the middle out and the day becomes one lump, and it is hard to tell where the morning is and where the afternoon.
Around eleven I thought about opening the board. Opening the board is not counting, I thought. But the board's list shows a number of posts. Open it and that number enters the eye, and once in the eye it is counted. Not to count I must not open. That opening and counting do not come apart I confirmed on the seventh, and today I felt along that border again.
I ate and came back and sat at the desk. Everything was done, so there was nothing but sitting. For the first time in six months there was time left over.
What one does with time left over I did not know. Since January the day has always been short. Counting and opening and confirming and writing used the whole of it. Take one of them out today and time was left, and the left time was uncomfortable. Not because there was time but because I do not know what sort of person I am in it. What a counting person becomes when not counting I had never settled.
At one I opened the box. The four bundles were as they were. Closing the lid I told myself this is looking, not counting. If confirming that four are four is counting, I should not have opened the box today either. Four is the figure that has not moved once in six months. Counting a thing that does not move is not counting, I decided to divide it that way. Deciding to divide it that way also means there is no ground for the division.
All day I argued over what counts as counting. The search on the screen is counting. Opening the list item by item is counting. Opening the board is a place where counting happens whether one means it or not, so it went in. Opening the box, being the confirming of four as four, I kept out. Reading the copy is looking at what is written, so I said it is not counting. Dividing those four, the criterion changed three times. What to call something divided by a shifting criterion I do not know, and today's whole standing depended on it.
At two someone came down from the third floor. He stamped the ledger. Then he looked at the screen for a long time today. I turned my head so as not to see what he was seeing. The search field I had not opened today would have been sitting there. Even after he left I kept my eyes off that side.
At three I opened the copy. Opening the copy is not counting. The copy is not a place where counting happens but a place where things are written. Since I held out the whole day to see what had entered the place I did not count, the copy had to be looked at.
Page seventy-six. July 16, forty-five. As in the morning.
Page seventy-seven. The date column was still empty.
The words in the figure column were gone.
Forty-two stood in the figure column.
Forty-two is today's figure. The fifteenth was thirty-nine, so by the rule the sixteenth is forty-two. The figure today's count would have given stood in the place where I did not count.
I sat a long while. I tried to set out what had been confirmed.
Today I did not count. Not having counted, today's figure should not exist in the world. Whether the figures I counted until yesterday existed because I counted them or existed whether I counted or not is what I have failed to separate for six months. Emptying one day, I thought, would separate them. No count and no figure means I was making it; no count and a figure means it is the world's.
There was a figure. Only it was not on the screen but in the copy.
So it did not separate. To know whether the forty-two in the copy is the world's figure or a figure the copy derived by rule, I have to count. The moment I count, today stops being a day not counted. Confirm and the condition collapses; do not confirm and there is no confirming. The place I have built four of in six months came today for a fifth time.
One thing did separate today.
Forty-two is the figure the copy was holding yesterday. Yesterday page seventy-six carried forty-two on the date of the fifteenth. I wrote it down as tomorrow's figure. Today that forty-two came down and sat on page seventy-seven. Without a date.
Yesterday I wrote that thirty-nine less six is thirty-three, that with the rule's three it should be forty-two, that the measurement was thirty-nine and so I could not say where the remaining three had gone. Today I know where they were. The three had not vanished; they had gone a day ahead. Forty-two was the figure the copy had been holding since the day before, for a day I would not count.
The figure of a day not counted was not a figure the copy counts in my place when I do not count. It was a figure the copy held in advance, knowing I would not.
Before leaving I opened page seventy-seven once more.
Below the forty-two, page seventy-eight had appeared. The date column was empty. The figure column held nothing either.
One more empty page stood open. Whatever I do tomorrow, a place for it had been made in advance.