April 14. One entry split in two.
At half past nine I counted up nine days' work.
When I scanned the four last January I read only the running prose at the front as logs. That ledgers follow on behind it I learned only now, copying them out. The page counts do not match the counts I made then either. Whether the ledgers have grown since or I undercounted then cannot be told apart.
The librarian's bundle is a list of holdings. A hundred and thirty pages, three or four titles to a page.
Each entry has a title and a number of volumes and a date received. Those three I copied out.
In nine days I copied all hundred and thirty pages.
For nine days I copied by hand.
I could have put it into the machine. The procedure does not say how to transcribe, so I did it by hand.
By hand it is about twenty minutes a page. Into the machine it is five or six.
There is no box for why I did it by hand. Put plainly, it was to read it. Into the machine the characters go past; by hand the characters stand.
The man seventy-six years ago wrote by hand too. He had no other way.
From ten to eleven I counted the entries.
I counted the original. Four hundred and twelve.
I counted the copy. Four hundred and thirteen.
From ten past eleven to twelve I looked for where the extra one is.
The same title stands on two rows.
The seventy-second and the three hundred and sixth.
The two rows have the same characters. The same number of volumes and the same date received. Only the place differs.
I opened the original. That title is there once. Page nineteen, third line.
There is nowhere to tell which of the copy's two rows is that line. Both are copies of the same thing and both are right.
At ten past twelve I looked at what the title is.
Six characters. One volume, received the sixteenth of December, twenty-four.
The sixteenth of December is a date this room has seen twice. There was one in ninety-seven and one in ninety-eight, and the two were inside one year.
It is in a list of seventy-six years ago too.
That makes three. The third is the oldest and it showed itself last.
At half past twelve I opened the back of the librarian's bundle.
On page a hundred and eighteen there is something he wrote.
It says that in counting, the same book was counted twice. It says it happened twenty-nine times.
What he did with the twenty-nine is written too.
He did not remove them. He says there were no grounds for removing them.
There is one more line under it.
It says there is nowhere to write whether they are duplicates or not.
From one to two I read that line several times.
Seventy-six years ago this man had nowhere to write it either.
He left twenty-nine. I met one today.
What he counted was books. What I counted was the lines he wrote.
His list is a copy of the books. Mine is a copy of that list.
There is one layer more. It splits once at each layer.
At ten past two I looked at where the two rows sit.
Seventy-two and three hundred and six. Two hundred and thirty-four apart.
In the original that title is on page nineteen. Page nineteen is somewhere between the sixty-first and the sixty-seventh entry.
Seventy-two is near it. Three hundred and six is not. It is about a hundred pages further on.
While I copied, that line crossed a hundred pages.
Of the nine days, I copied page nineteen on the sixth. I wrote the three hundred and sixth row on the eighth.
Two days apart I wrote the same line twice. Writing it the second time, I thought it was the first.
At twenty past two I counted again.
Four hundred and twelve in the original. Four hundred and thirteen in the copy.
Counted twice it is the same. It has not grown.
What split is one, and that one grows no further.
From twenty to three to half past three I wrote down which side split.
The original is one and the copy is two. The side that split is the copy.
For eight months in this room what diverged was always the copy. The copy wrote a day early, the copy had one line more, the copy's number was larger.
Today it is the copy again.
One thing is different. Until now the copy was made by the machine. Today's copy was made by me.
At twenty to four I decided what to do with the two rows.
Remove one and the copy agrees with the original. There are no grounds to remove it. Both are copies of what is in the original.
Leave them and the copy has one more.
I took the side the man seventy-six years ago took.
I wrote the reason in the margin. The margin entries this month are four.
At four I made the table of contents for sub-number one.
It takes the first and last title of each page. A hundred and thirty pages, so a hundred and thirty lines.
That title goes on the line for page nineteen. In the copy it is two rows, but the table runs by page, so it catches once.
In the table the split cannot be seen.
What is not in the original is in the copy, and in the table made by reducing the copy it is gone again.
What increased vanishes in the reducing. What vanished is still in the copy.
At ten past four I opened the right-hand material.
Three hundred and twenty-two pages. The date of the added page is 13 May 98.
My log is on its three hundred and twenty-second page today.
At half past four I opened the box.
Seven. Six impressions. The fiftieth day.
At five I looked at the sixth box.
Five hundred and twenty-one.
At ten past five I wrote the ten-line summary for sub-number one.
A hundred and thirty pages, four hundred and twelve entries. Of twenty-four, recording what came in over one year. Title, volumes, date received.
I read the ten lines again.
The split title is not in those ten lines.
It is one of four hundred and twelve and there is no room for it in ten lines.
Whoever opens the set sees ten lines.
In ten lines nothing has split.
Reducing it, what split goes first. What split is one of three, and ten lines is one of seventeen.
That there is no box for what is dropped was written first by a man seventy-six years ago.
He wrote it at the end of his list. The list remained and that line remained with it.
Where I wrote it today is a margin.
A margin does not go into a summary.