Multiple-choice reading

For the first few years of my working life, I thought of silence as a kind of raw material, something to be found, captured and sold. I recorded wildlife sound for broadcasters and, increasingly, for the makers of relaxation apps, and the brief was nearly always the same: birdsong, running water, wind in trees, and nothing else. Nothing else meant no aircraft, no traffic, no distant chainsaw, no farm dog. It meant, in practice, a version of the countryside that hardly exists any more, and my job was to manufacture the illusion that it did. I was good at it. I knew which valleys lay outside the flight paths, which hours of the morning fell between the milk tanker and the school run, and how to edit out a passing tractor so cleanly that no one would ever know it had been there.

The doubt crept in slowly. I remember a dawn in the hills of mid-Wales when I had been waiting for nearly three hours for a gap in the noise long enough to record a curlew. Each time the bird began to call, an aircraft would pass high overhead, inaudible to most people but perfectly clear on my headphones. By the time I got a usable minute, the sun was up and I was too cold to feel pleased. Listening back that evening, I realised that the recording I had fought for was, strictly speaking, a lie. The curlew had never sung in silence. It had sung in a world full of engines, and that world was the one it was struggling to survive in.

It would be convenient to say that I gave up commercial work on the spot. I did not; I had a mortgage. What changed was that I began, alongside the clean recordings, to make another set, the ones I had always thrown away. I kept the aircraft and the tractors. I left the motorway hum running under the skylarks. At first I did it out of a vague sense of duty, as if someone ought to keep a record of what places actually sounded like. Later I found that I preferred listening to these recordings. They were less beautiful, but they were truer, and there was something in the way the birds pitched their songs around the noise, finding the gaps, that moved me more than any untroubled chorus.

Scientists, it turns out, have been listening to the same thing. Studies in several countries have found that birds living near busy roads tend to sing at a higher pitch than their rural counterparts, presumably to make themselves heard above the low rumble of traffic. Some sing earlier in the morning, before the rush hour begins; others simply sing louder. None of this is free. A bird that must shout to be heard spends energy it might otherwise devote to feeding its young, and a song that has been altered to cut through noise may be less attractive to a potential mate. The adaptations are ingenious, but they are also a measure of strain.

When I play my 'honest' recordings to audiences, the reaction is often one of mild embarrassment, as though I had shown them a photograph of a beauty spot with the car park left in. People want the clean version. I understand that; I wanted it too, for years. But I have come to think that the clean version does a quiet sort of harm. If the only countryside we ever hear is one without engines, we lose the ability to notice how much has changed, and the loss of quiet, unlike the loss of a hedgerow or a woodland, leaves nothing visible behind. No one campaigns for a sound they have forgotten was there.

I still sell clean recordings, and I am not sure I could defend every one of them. What I try to do now is to be candid about how they were made: the three hours of waiting, the edits, the aircraft quietly removed. It seems a small thing. But a listener who knows that a minute of birdsong took a morning to capture is, I like to think, a listener who has begun to understand what the morning was like.