Marcus Webb kept records the way other men kept faith, not because he expected to be asked for them, but because something unwritten was something he could not later trust had happened at all.
In the spring of 1985 he was thirty-two and worked out of two rooms off Menaul Boulevard, which was a generous description of both the work and the rooms. Menaul was a wide, plain commercial street on the Albuquerque flats. He had left the national laboratory four years earlier without a grievance and without a plan, which the people he left behind had found more suspicious than a grievance would have been. Now he characterized materials for whoever paid him to. A company would send a sample and a narrow question: how something failed, and at what temperature it began to. He would send back a number and an invoice. He was good at it in the unglamorous way that made a man useful and easy to forget, and he had come to suspect the two were the same quality, measured from different distances.
The contract came in April, on paper with no company at the top of it.
He noticed that first, the way other people noticed a face. There was a scope of work and a rate that was wrong by a factor he counted the zeros of twice. The scope described an object without saying where it was from, or what anyone believed it to be, or what he was expected to find. He was to determine how it behaved. Its mechanism. Nothing else. The origin, a later clause explained in the flat voice of something already decided by people he would never meet, was not within scope. He read that sentence four times. It was the first sentence he had ever been paid not to be curious about.
The man who brought the paperwork came once.
Webb tried, in the years afterward, to write the man down, and could not. That failure sat worse with him than anything the man had said, because Webb did not fail to write things down. He could recall the office, the light through the blinds at four in the afternoon, the exact weight of the envelope. The man himself came back only as a shape and a courtesy. No name that stayed. A voice with no particular place in it. What stayed, in the end, was not the man at all but the ash he had left on the floor by the chair, from an off-brand cigarette Webb had not once seen him raise to his mouth, and the small indignity of having knelt to clean it up. He said the work would matter, and that it would stay quiet, and that Webb had been selected.
"Why me," Webb said. He gestured at the two rooms, at the secondhand equipment, at himself. "You understand I'm nobody in particular."
The man considered this as though Webb had confirmed something rather than asked it.
"Yes," he said. "That's right."
Webb was thirty-two, and he heard it as candor: a plain acknowledgment that discretion was worth more here than reputation, that a man with nothing to protect was a man who could be trusted with something. He was even, in a small way he would not have admitted, flattered. It would be a long time before he let himself hear the other part that had been in it: that nobody in particular was not a limitation they were prepared to overlook. It was the specification. They had not chosen him in spite of it.
The object arrived eleven days later, under a different arrangement, in a crate built for it and nothing else. The men who brought it set it on a cradle in the inner room and left without uncovering it.
He was not sure what he had expected. Something ugly and Soviet, perhaps, or something so ordinary it would be an insult to the rate. What he found under the shroud was neither. It stood taller than he did, and enclosed a space at its center about the size of a doorway, and it was heavier, he would learn, than anything that size had any right to be. He stood back from it for a while, under the good light, and simply looked, the way he had not looked at anything since he was a boy.
It gave up nothing to the eye. No seam whose start he could find. No mark, no maker, no up or down that the shape insisted on. It was the most finished object he had ever seen, and standing there in the four o'clock light, with a legal pad he had not yet written a word on, he understood that it had been made. A finish like that did not occur; it was achieved. He had been hired to explain how, and told, in writing, not to wonder by whom.
He picked up the pencil. He wrote that day's date at the top of the page, because that was where a record began, and because the date was the one fact about the object he was certain of.
Then he began to measure it, not knowing what it was.
He would not know for years. And there was a smaller thing besides, the one he set down that afternoon and did not pick back up for a decade, that he would never entirely know at all: not what the object was, but why, of all the people who could have been given it, it had been given to him.
For two years the object defeated him, and it did so politely.
It was not that it hid. It answered every question he put to it, and answered the same way twice, which in his trade was the definition of a fact. The trouble was that the answers did not belong to the same object. Its density, worked out from its volume and its weight, came to a figure no material he knew of reached, or even approached. He checked the scale against known masses. He checked his volume against water. The figure held, and meant nothing to him, the way a correctly spelled word in a made-up language means nothing to anyone but the people who made it up.
He might have written it off as some dense alloy he had not been cleared to identify, and for a while he did. What he could not write off, in the second year, was the other weight.
He had needed to shift it once, only a little, only to get a cradle arm under it, and had rigged a jack and a winch to do it. It had not needed either of them. It moved at the push of a hand, drifted like something a tenth its weight, and settled again the moment he stopped, while the scale beneath it went on insisting it weighed exactly what it weighed. He did not trust the observation. He built a way to make a number of it. He measured what the object weighed, which was one quantity, and then he measured how hard it was to move, which by every law he had been taught was the same quantity in other clothes, and the two numbers were not the same. They were not close. He ran it eleven times across three days and got the same disagreement to the third decimal, and on the third day he sat down on the stool by the bench and understood that he had measured something that was not allowed.
The breakthrough, when it came that autumn, was not a discovery. It was a decision to stop defending the physics and start believing the object. He remembered later that it had been late, and that he had not been working when it happened but standing at the window with a cup of coffee going cold, looking at nothing, the way answers seem to prefer.
If the thing it weighed and the thing it resisted were different, then the object was not, in the sense he had been treating it, a material at all. It was a machine, and its purpose was the discrepancy. It held the relation between weight and mass slightly open, the way a finger holds a door, and everything strange about it followed from that one held door: the density that was no density, the finish that no process reached. He did not have the whole of how. He had, by the end of 1987, a partial account, a set of equations that described the mechanism without quite closing around it, and they were beautiful in the specific way that meant they were probably true.
He sat with the pages for a long time. He had solved, or half-solved, the most important thing he would ever touch, and there was no one in the two rooms to tell, which was correct, and which he minded more than he had expected to.
He assumed, because it was 1987 and because a man assumes the world he lives in, that somewhere a program he would never be read into had built this, and that his part had been to work out what they already knew and could not write down for reasons of their own. It was a comfortable assumption. It lasted as long as it took him to ask the next ordinary question, which was what the mechanism was made of, because a machine that holds a door open has to be made of something that can stand the strain.
Nothing could. Nothing he could name, and nothing the equations would allow to exist at the temperatures and pressures of anywhere he had heard of anything being built. The account of the mechanism was clean. The account of its manufacture was blank, and stayed blank, and a blank that would not fill was something Webb had no procedure for.
He wrote the report the contract had bought.
It was a good report and an honest one, in the narrow way he had been paid to be honest. It described the mechanism to the edge of what he could defend and not past it. It used the word gravity, which was the true word, and none of the words that stood behind it, because he could not defend those and did not, yet, let himself think them. The other pages, the ones about what the object was made of and could not have been made by, he kept in a folder that was not part of the report, in a drawer that was his own, and told no one they existed.
He was thirty-four now. It was the first thing he ever knew that he decided not to share, and it went into the dark of the drawer easily, the way the first one always does.