August 22, 2015

A change of tune

In our garden, along our street and on the common our local robins are coming out of moult and starting to sing again. Because so many other birds are now silent, or preparing to fly south, their song stands out: that sad, descending query, always ending unresolved, trickling from tree and thicket like a premonition of autumn. Walking the dog today, we found ourselves standing for a moment, searching out the tiny, feathered body with its orange bib from which the notes were pouring out. It is lovely, after the August hiatus, to hear them again.

People say that robins sound more pensive in autumn and winter than in spring, and maybe that's true; or perhaps in April and May we merely ascribe to them the joyfulness we believe we hear in other birds' songs, like the blackbird's; or even the optimism we feel ourselves at the start of the year. August is one of the year's fulcrums, and as the first leaves start to turn – no matter how warm the weather – we feel the world tip inexorably towards winter. No more growth, or flowering, or seeding: instead, a falling-back. Little wonder the robin's song sounds now like a lament.

It's not just the season that causes a variation in robins' songs. Today we heard one – perhaps a youngster trying out his voice – that sounded very unusual, far more beepy and mechanistic than they usually do. Another was letting out a high-pitched "whee, whee, whee" which I wouldn't have identified as being from a robin had I not seen it doing it. Then there are their "dit!dit!dit!" alarm calls, and their contact calls, too. It only takes a few minutes on birdsong website Xeno-Canto to realise the huge variety of vocalisations they make.

And this is a big part of the reason I can't get behind Chirpomatic or Warblr, two birdsong recognition apps being touted as 'Shazam for birds'. The dream is to develop software that will allow anyone with a smartphone to find out what the bird is that they can hear singing – but add distance, wind noise, background sound, other birds' calls and the limitations of phone mics and processors to the natural variation I've already described and it's obvious that the dream isn't very achievable. Our brains may be excellent at 'fuzzy logic', but computers are hampered by Boolean. I was not at all surprised to read that the first results from these apps haven't been good.

I feel strongly that in their current state these apps are, in fact, worse than useless, because they will encode errors that will then become hard for beginners to shift. If it tells you a nuthatch's call is in fact a great tit (a bird which, by the way, is thought to have over 40 different types of call), you'll take that mistake away with you and carry on misdiagnosing nuthatches. Without the app you'd mark the song down as 'unknown' and continue to seek visual confirmation (which is all it takes to be sure); but when technology gives you information, the risk is it becomes fact in your mind.

And there's more to my misgivings, too. Why must we always look for short-cuts and quick fixes; why must we turn to technology instead of trusting our own ability to learn? A few years ago I only knew a few birdsongs; now I can recognise nearly all the birds I'm likely to hear in my local area – though take me out of town and I often feel like a beginner again. It wasn't hard, it just took tuning in: stopping, looking and listening, as we do in childhood. Anyone can do it. Most importantly, there was a joy in doing it, an immense satisfaction in getting to know my avian neighbours, and gradually improving my ear. There still is.

I'm not against using apps about the natural world; far from it. Being able to carry reference books around with us in our pockets is a wonderful thing, as is the ability to use our phones to record sightings and share knowledge. But Warblr and Chirpomatic ask us to outsource to technology a task we can do ourselves – that of listening – and in doing so, they do us a disservice, preventing us from learning to hear the subtly different types of song.

We are not yet so removed from nature that we need an app to make unreliable guesses at which birds are singing around us: our own curiosity and patience, our  ability to learn and connect and remember, is everything we need.



Addendum: The people behind ChirpOmatic got in touch to say this: "We absolutely agree with everything you say, which is why ChirpOMatic is designed as a learning tool, NOT a shortcut. You record, it offers choices, you read descriptions, listen to choices and choose. Your recording is stored to use as reference. It works in conjunction with our Chirp! app which gives more info, more sounds, quizzes – all to help you learn."
Further, Florence Wilkinson of Warblr sent me an email, which you can read by clicking here.