The Copier and the Predecessor's Number

That day's work, too, ended with the copying of old ledgers onto microfilm. The records room was on the second basement floor of the merchant bank's main building, and one panel of the fluorescent lights always flickered. I sat at the camera stand, spread one leaf of a yellowed ledger under the glass plate, and pressed the shutter. The film wound on one frame with a sound. The next leaf. The shutter. The next leaf. There was a kind of solace in the copying — the numbers stayed in their places, and when I copied a leaf, that leaf remained in that place forever.

What a records clerk does at a merchant bank is to copy what will vanish into what does not vanish. Ledgers past their closing, notes past their maturity, accounts settled and closed. Head office sent them down in a can of film to the storeroom, and in between I copied the numbers. It was the autumn of 1997. In those days the people upstairs spoke in low voices in the corridors of foreign currency, and of short-term. I had no occasion to copy those words. My work was the ledgers already finished — until then, at least.

Before me stood the desk my predecessor had left. Who he was, I do not know. The handover had been finished by the personnel office with a single sheet of paper, and all he left behind was a bundle of index cards half-sorted and a camera stand whose focus knob was worn from being left on too long. One day he did not come to work, and the personnel office processed the post as a resignation. The letter of resignation was not preserved.

That afternoon, a box came down to the records room from the main storeroom. The materials of a defunct client, the conveyance slip said. The client's name was a publisher — 'Pnakotic Press (Pnakotic 文庫)'. What dealings a merchant bank had had with a publisher, the slip did not say. Only that the publisher had gone out of business with its debt to our company unsettled, and that the materials pledged as collateral had drifted into the storeroom — this I learned later.

The box was bound with cord, and the knot of the cord bore the grain of a hand that had often touched it. I untied it. Inside were four bundles of records. Not printed books, but logs written by hand. One a lighthouse keeper's, one a surveyor's, one a wireless operator's, one a librarian's. On the cover of each bundle was stamped the same press's seal, and the same single line.

I traced that line with my fingertip. A cold like the chill of the basement ran down my spine. "The basement in October is cold like this as a matter of course," I muttered to myself, trying to give a reasonable account of the chill that seemed to pour from the box. It was not that the box was cold because it sat there — only that the box had been set in a place already cold, I added.

It was then that I remembered my predecessor's bundle of index cards. Those cards, half-sorted. I drew them out and looked at the last one — at what material he had last been indexing. In the classification line of the card was written the familiar name of the press.

Copy it down, but do not read to the end. It was my predecessor's hand. An indexing clerk does not, in sorting his materials, write "do not read." For an index is a thing one must read to be able to write. I tried to give a reasonable account of that one line, but the account did not arrive.

The last work of the day was to copy onto film one closing ledger head office had sent down. It was the ledger of a settled account. I spread one leaf under the glass plate and pressed the shutter. It was the leaf on which the balance of a short-term foreign-currency borrowing was written. A number copied out neatly to its last figure. I committed it to film, and copied the balance by hand onto an index card. The number did not lie. It was a settled balance of an already-finished ledger.

I set down the day's record so and turned off the lights of the records room. The fluorescent panel that had flickered flickered once more, last of all, and went out.

The next morning, I collated the short-term foreign balance I had copied the day before against the ledger. The frame on the film was as it had been the day before. The number I had copied by hand onto the index card was as it had been. But the balance on that leaf of the ledger — the leaf settled and closed, that would never be written again — had grown by one figure at the end.

I looked at the grown last figure three times over.

The short-term foreign borrowing balance I had copied the day before. The frame on the film was as it had been; the number I had written by hand onto the index card was as it had been. The two agreed with each other. What did not agree was the ledger. The balance on that one leaf — settled, closed, never to be written again — alone bore one more figure at its end.

I tried to give a reasonable account of it. Someone had cut a correction slip and revised the balance — even for a closed account, an after-the-fact correction is possible. I opened the corrections register and found the account number. On that leaf there was no correction. That day, the day before, no one had touched the account.

To believe it was all, I needed another record to compare — the same account, copied by someone before me.

I drew out my predecessor's bundle of index cards again. Those half-sorted cards. Yesterday I had looked only at the last one; today I turned them from the start. He had been a meticulous man. One card per account — balance, date, even the number of the film can. The card for the short-term foreign borrowing was near the middle of the bundle.

He had copied that balance in July of that year. Four months ago.

And the number he had entered in July was the very number the ledger had reached this morning, having grown by one figure.

I brought the card close to the lamp. I had not misread. The short-term foreign balance my predecessor had copied by hand four months ago, and the balance the settled ledger had grown to of its own accord this morning, were the same down to the last figure. The ledger had not grown. The ledger had climbed — belatedly — toward a number someone had already set down four months before.

The chill of the basement rose into my fingertips. I tried to give a reasonable account of it, but the account did not arrive.

Beneath the balance field, my predecessor had written one line in small characters. But the line had been struck out. Not with pencil — scraped away with a blade's edge, the grain of the paper lifted. I tilted the card aslant to the light. Through the scraped place, the outline of the pressed-in writing rose.

The next card's number. I weighed what it meant. In July, he had set down first the number of the next card not yet written. The number the settled ledger took four months to climb toward. He knew where it came from — and so he scraped it away.

The one line he had left on the last card now read differently.

I had read it as a charge concerning the four bundles from Pnakotic Press. Do not read the old logs to the end. But what he had been copying was not logs — it was ledgers. What he warned not to read to the end was the numbers. The figure that fills one cell and runs on below it, and below that.

I looked again at the short-term foreign borrowing card. He had set down the balance, and the film can's number. And in one corner of the card, with an arrow, he had written the number of the next card. It was his habit, threading the index. The number the arrow pointed to should have been in the next slot of the bundle.

I searched for the card with that number.

It was not there.

The next card the arrow pointed to was nowhere in the bundle. In its place was only a gap the width of one missing card. Someone had drawn it out — or the card meant to be written had never been written.

I laid my finger in the empty slot. The gap, one card thick, was cold. Distinctly colder than the cards beside it.

So long as that slot stayed empty, the arrow had nowhere to go. But by my predecessor's habit, that next card too would have borne an arrow. Pointing on to the next. The number ran on from card to card, pointing to its next before it was even written.

That day I could not turn off the lights of the records room. Where the next card from the empty slot was, I had already begun to guess. In the very can my predecessor had last been winding through the camera. In the film, on the day he did not come to work, that had never been wound to its end.