The radio telescope on the far side of the hill is a dish on a mount that points where the schedule says and records what it hears, and the schedule is mine. Every night at half past three it leaves the schedule. It slews to a point in the sky, tracks it for eleven minutes, and records, on a frequency the receiver is not tuned to, a signal. The signal, played as sound, is a woman's voice, counting, slowly, from one, and she gets to six hundred and sixty, and stops.
I am the operator of a small radio telescope at a university observatory, and the dish is mine from dusk to dawn, except for eleven minutes.
The dish.
Fifteen metres, on a mount, on the far side of the hill from the optical domes, pointed by a schedule I write, observing at frequencies the receiver is built for, recording everything to disk. It is a dull instrument doing dull science and I love it.
At half past three it leaves my schedule.
The slew.
The log shows it. Scheduled target dropped. Slew, to coordinates that are not in the schedule, a point in the sky with nothing at it in any catalogue I have. Track, eleven minutes. Record. Then back to the schedule as if nothing had happened.
The recording is at a frequency the receiver is not tuned to. It is a frequency the receiver cannot tune to. It is recorded anyway, as a file, and the file has a signal in it, narrow, strong, and when you play a radio signal as sound, which you can, it is a voice.
The voice.
A woman. Counting. One. Two. Three. A second apart, steadily, the way you count when you are timing something and have nothing but yourself to time it with. She gets to six hundred and sixty. Eleven minutes. And stops, and the dish goes back to the schedule.
I have listened to a year of them. The voice does not change. The count is always six hundred and sixty. And at about four hundred, every night, there is a break in the counting of about two seconds, as if she has looked up from what she is doing, and then she goes on from where she was.
What has been checked.
The receiver, the correlator, the control system, the lot, by the engineers, who found no way the dish could be slewed by anything but the schedule and no way the receiver could record at that frequency, and who listened to the file and left.
The dish, at half past three, by me, at the base of it, in the field. It slewed. I heard the motors. It came round and up and stopped, pointing at a part of the sky I could see, and there was nothing there but stars, and it tracked, and in the control hut the recording started, and I could hear it on the monitor speaker, the counting, and I stood under fifteen metres of steel pointed at nothing while a woman counted to six hundred and sixty from a place in the sky where there is nothing to count from.
At four hundred she paused. I looked up, because she had. There was nothing.
Who.
The dish is not old. The hill is. Before the dish there was a hut on the far side of the hill where a woman, in the sixties, did the observatory's timing work, which was a real job then: she kept the clocks, and she timed occultations, which is when the moon goes in front of a star and the star goes out, and you time how long it is gone, by counting, if you have nothing better, and the observatory's records have her timings, pages of them, in her hand, and they are good.
Her last page is a night in 1969. An occultation, predicted to last about eleven minutes. The count is there, in her hand, to four hundred and some, and then the page stops, and the department's history says she had a stroke at her desk that night, in the hut, alone, and was found in the morning, and the timing was never completed, and it had been, the history says, an important one.
She is completing it. Every night. From the point in the sky where the star was, at the frequency of nothing, counting to the end of the eleven minutes, and at four hundred, where she stopped, she pauses, and she goes on.
What is done about it.
The eleven minutes a night are on the instrument log as an unscheduled slew and lost observing time, and the time allocation committee has asked me to account for it, and I have said that the instrument is completing a historic observation, which is what I wrote on the form, and they have asked me to explain, and I have sent them the file.
What I do.
I operate the dish.
A timing is no use if nobody writes it down. She counted and nobody wrote it. So at half past three I am in the hut with her page, the 1969 one, which I have had out of the archive, and a pencil, and when she reaches four hundred and pauses, I write the next number, and the next, in the column under her hand, in mine, and I keep writing until she says six hundred and sixty and stops, and I write that, and I write the end time, and I sign it.
The page is complete. It has been complete since the first night I did it. The timing is in the archive with her name and mine, and I sent it to the people who keep such things, and they wrote back that it agreed with the prediction to within a second, which it would, and thanked her.
She counts every night anyway. I think she does not know. But the pause at four hundred is shorter, now, a second, less, and I think that one night she will not pause at all, and will count straight through to six hundred and sixty, and that will be the night the dish stays on my schedule, and I am going to miss her, and I have told the committee so, and they have stopped writing.