December 16. In the morning I sat down to count how many of this date I have.
On the eighth of November the cards ended.
I wrote four digits beside the last card of the twelfth drawer and closed the drawer. Having closed it I counted, and there were seven thousand three hundred and sixty-two. What I reckoned in January was seven thousand three hundred and forty. Twenty-two cards are outside the reckoning.
In October I wrote that it was work with an end. It ended. Not by the reckoning.
Thirty-seven days have passed since.
In that time there was work in taking requisitions, bringing out boxes and entering them in the register. That work takes a hand. Only there is nothing to increase. There is no place to correct and no paper to move. In the morning I set the search running and write the number down, and after that I read.
The work I do in this room is reading and nothing else.
In the thirty-seven days I read the fifth bundle three times.
It is a hundred and seventy-six pages. The numbers written at the head of the entries stop at sixty-six. There were many days she filled several sheets, so counted as paper it runs far longer. The first page is the twenty-seventh of October 97, and the box came that same day. After receiving the box, what that person did was copy out the four inside it. It is written down that the company did not order it. What was ordered was moving the old ledgers onto film, and the four were done in the gaps.
There are thirty-one places where the number copied grows larger than the original. Three counts or four counts, and each time it is written that all three came out larger the same way. Nowhere is a reason for the disagreement written. In the place for a reason there is a line saying what will be tried next.
I counted how many places are the same as what I have seen in this room in eight months. Nineteen. There is a place the same as the copy writing a day first, and a place the same as a date not in the catalogue showing first. That side is paper and this side is a screen. Paper and screen differ and the places are the same.
Nowhere in a hundred and seventy-six pages is it written what this is. There are some hundred questions and nowhere an answer.
Having written nineteen down I meant to verify one thing. Whether the same thing repeats across two years, or whether one set follows the other.
If it follows, there must be a which is first. By date that side is two years first, and my reading of it is this month. It is not that they became the same after the reading; they were already the same before it.
That they were the same before the reading I cannot verify. There is a log with eight months in it, so I can verify it. That log was written by my own hand, so it cannot be verified.
The word for what I do is one I found in July. Beneath the last page of each of the four bundles all four carry the same one line.
We are watching a little further.
On the twenty-eighth of November the head office clearing guidance came.
It said to clear that line dated the thirty-first of October as material absent. I cleared it. The line is not deleted. A clearing mark is attached as one line beneath it.
The lines standing under that material number became five. In October there were four.
I followed guidance that said to correct, and one was added. I was not surprised. That I was not surprised I wrote in the log that day.
From half past nine until eleven I counted the sixteenth of December.
First, the displayed end date of the right-hand material is the sixteenth of December 97. That mark has been there eleven months.
Second, my predecessor's date of separation is the sixteenth of December 98. It is written on the first page of the handover papers received in January and I have looked at it many times since. The notebook's last writing bears that date too.
Third, today is the sixteenth of December 99.
Having written three, I opened the box.
The date of the fifth bundle's last written page is the sixteenth of December. The year box is empty. By the way I have counted for eleven months it goes into none of the three. It is a fourth.
At one I switched on the test machine.
In October I rolled the clock over and did not roll it back. Rolling it back takes half a day, and once it is back there will be testing to do again. So that machine lives in year double zero and runs a separate volume of its own log.
I queried today's date.
It came out as the sixteenth of December, nineteen hundred.
That is a fifth.
I looked at the sort. Today's stands at the top of the list. Nineteen hundred is earlier than all four of the four bundles' years, so in the inverted list today is the first line. In October I saw that for the first time and wrote that in six months I had taken the top box for the first time. It is the second month in that place.
I sat down to count three and wrote five.
That counting is increasing I learned in this room. I learned it so with the query count, so with the cards, so with the date fields. What increased was always a thing or a number.
What increased today is a date.
A date is not a thing held in this room. It is not paper and it is not a file. It is what gets written on paper and in files. Whether the places it is written increasing means the date increased or the places increased, I could not tell.
From half past one until two I set the five out and looked at them.
Three of the five are days a position closed. On the sixteenth of 97 the desk beside the right-hand person went empty, and that person's log received its end mark under that date. On the sixteenth of 98 my predecessor was separated. On the yearless sixteenth the writing of the fifth bundle stopped.
The fourth is nineteen hundred, and in that year this room did not exist.
The fifth is today.
At half past two the third-floor man came down and took the box. He asked whether it was quiet, being year end. I said it was quiet. He said he heard I was on night duty on the thirty-first and that it would be hard, and went up.
The circular came at three.
It is an order for night duty on the thirty-first of December. It says one person from the computer room, and in the name box my name was written.
Since taking the post in January, no head office circular has come with my name already in it. They came to the department and they came by employee number.
The third-floor man knew at half past two and the circular came at three.
At half past three I opened the right-hand material.
It was two hundred and eleven pages. The page added is dated the twenty-second of January 98. The displayed end date is still the sixteenth of December 97. It has gone thirty-seven days past.
A log that received an end mark has written thirty-seven days more.
From four until five I read those thirty-seven pages. It is January 98. What this company did that month is in the paper register. In that log there is what was copied and how many were counted and how many did not agree.
Inside the boxes is to the form and outside them sentences are mixed in. Six in thirty-seven pages.
All six are questions.
One of the six I found in the fifth bundle. On page one hundred and twelve of the bundle the same question stands in prose. On the log side it is one line outside a box and on the bundle side six more lines are attached before and after.
One person wrote two sets. Where there were boxes one line, and where there were none, seven.
The boxes did not take that person's six lines. My boxes did not take my nine lines either, and I held that this was keeping the form.
At half past five I took out the fifth bundle.
For thirty-seven days it had not grown past a hundred and seventy-six pages. Things that do not grow are rare in this room, so I counted it every day.
Today it was a hundred and seventy-seven.
I opened the page added. The date is the sixteenth of December.
Two digits had come into the year box.
They were double zero.