Part 4: ‘I bird with Bill Smith’

No bird has just one call or one song

As you can tell from the quotes and references in this guide I am a bird expert groupie. It’s not all that I am, but it was the point made to me by my friend Steve Smith when for my birthday he gave me a T shirt with ‘I bird with Steve Smith’ printed on it. T shirts are a bit like an advert or label and I have to admit to never having worn it. Anyway I’ve now found a bird expert called Smith, although I’ve never met him. W J Smith isn’t fond of labels either and he wrote a vigorous attack on the oversimplified labelling of bird calls in our field guides (Smith 1996) that has encouraged me to think twice before labelling a bird sound.

It seems reassuring and straightforward to label a sound as an alarm call, a flight call, territorial song and so on, but Smith emphasised that birds use the same vocalisations for more than one purpose or strategy; one and the same signal can convey different messages depending on the recipient. So when we are tempted to interpret the meaning of a sound, we need to be as broad-minded as we can.

It’s easy to misinterpret or oversimplify a bird’s behaviour and there is a great story that illustrates this point in Mind of the raven by Bernd Heinrich (2000), where the author tells of a man who was being chased by a Grizzly Bear Ursus arctos horribilis. As the man ran, a Common Raven Corvus corax guided him with its calls, and by following this bird he miraculously escaped the bear. On recounting the story to the author, a renowned expert on raven behaviour, the chap was shocked as it was explained to him that the raven was unlikely to be guiding him but rather calling the bear to the man so that it would kill him and enable the raven to feed off the scraps.

Looking at identification papers in most journals, or in field guides, the convention to name the song or the call is powerful and widespread. Yet read The birds of the Western Palearctic (BWP) or an original research paper and you may be amazed how many vocalisations can be attributed to a Common Blackbird or a House Sparrow Passer domesticus or indeed a Common Raven. The misconception that each bird has just one song and one call is only a slight advance on childhood, when we allowed them one sound each: all ducks said quack and all owls said too-whit too-woo. In its adult guise, this kind of thinking is responsible for much of what we seem to screen out. Near where I live in Dorset is the swannery at Abbotsbury. When I first heard the variety of sounds made by Mute Swans as they were being fed, I was really surprised. I thought they were supposed to be mute, and yet even these birds have a range of different vocal and hissing sounds, besides the wing sounds and grunting call we heard earlier (CD1-08). Slowly over the years I noticed Mute Swans calling everywhere. I don’t know what species I had attributed these sounds to before. Common Moorhen Gallinula chloropus? Eurasian Coot Fulica atra? It’s serendipity I suppose, as with so many subjects; the more you read and listen, the more variety you will discover.

There can often be occasions when it is difficult to know where to draw the line between a ‘song’ and a ‘call’, and it is best not to be too rigid in making those distinctions. In early June 2002, I was in Finland listening to a Brambling F montifringilla (CD1- 62), and as expected it was singing a flat buzzing note repeated every five seconds or so. Later that month, while sitting on the wall by the river outside my hotel in Gavarnie high in the French Pyrenees, I recorded a Common Chaffinch perched on top of a small tree making a similar but shorter sound and delivered faster (CD1-63). I started to think about Smith’s warnings. This kind of sound is usually labelled as a call, and in fact a whole class of repetitive sounds made by chaffinches from prominent perches are known as ‘rain calls’. In any given area, all birds will use the same note, and these sounds are traditionally considered to show regional dialects as they vary from location to location. Whatever sound is used, a buzz, a whistle or a more complex sound, it is always delivered at regular intervals. 

Smith’s attack went like this: “Field guides recognize as songs only the most widely accepted cases… Other vocalisations are listed, if at all, as nonsong ‘calls’. The simpler vocalisations of many species are relegated to nonsong status even if often uttered in the sustained quasi-rhythmic bouts diagnostic of singing performances” (Smith 1996). Smith was talking of North American field guides but in fact most field guides, for brevity and convenience, have perpetuated the myth that birds have one song and one call, and rarely question what belongs in which category. Once you consider chaffinch ‘rain calls’ from Smith’s perspective, you have to question whether they make more sense as songs. 

Having read more on the subject, I now realise that my thoughts on rain calls were not original. Thielcke (1969) was of the opinion that the rain call was a kind of a “song substitute”. The lack of a musical comparative chaffinch-like song in Brambling is intriguing, although if both taxa have two choices of song type it makes perfect sense for the taxa breeding in the colder north to drop the complex song and opt for an energy saving repeated call-song. So where does this leave us? How should these sounds be described: song or call? Draw your own conclusions as you listen to the Brambling and Common Chaffinch examples, then listen to a recording of a typical European Greenfinch Chloris chloris, which is also repeating a commonly heard Brambling-like sound in a song-like performance (CD1-64). 

Short-billed Limnodromus griseus and Long-billed Dowitchers L scolopaceus are a classic pair of birds in which sound clinches identification. Travelling in Texas with Anthony and Bruce, I spent a lot of time asking them to describe what they could hear when either species flushed. Bruce had very little experience with Long-billed, whereas for Anthony it was Short-billed that was less familiar. One time on San Padre Island we were discussing the finer details of separating them, when Anthony suggested flushing a small group of dowitchers just beyond a reedy fringe. Five alligators noisily broke cover in front of him. I listened to my tape of the event recently and “alligator infested” was all he said, then confirmed the identification as “Long-billed”. When things calmed down we started to realise that the call they make when they take off is not the same as the one they use in steady flight. If you are ever lucky enough to find a dowitcher Limnodromus in Europe, you’ll really need to hear ‘the’ call. Imagine the scene (as often as I have); looking round to make sure no one is watching you flush the bird, and fortunately it takes flight giving a short series of calls. Checking the American field guide you bought to keep in the pannier of your bike, just in case, you read “… flight call … Short-billed Dowitcher … a staccato trebled tututu;
… Long-billed Dowitcher … a single thin keek occasionally doubled or trebled” (Peterson 1980). This was what happened to Irish birders in the sixties who heard dowitchers giving multiple-note calls as they took off and thought, reasonably, that they were hearing Short-billeds. Several were identified in this way. Now we are better informed on the matter of dowitcher vocalisations. We know that when Long-billed take flight they often give a rapid series of short notes, rather like Short-billed. The diagnostic high-pitched, single-note keek is used both on the ground and when the bird is in steady flight. The written descriptions of calls in accounts of the Irish dowitchers were excellent, enabling those reviewing the records in later years to correctly record the birds as Long-billed. 

It’s a strange aspect of a birder’s psychology that having found a rare bird he returns to the same site over and over again in subsequent years partly paying homage to “The Sender” for the blessing and partly in the hope of seeing the bird again. On the last day of June 2004 Killian was doing just this. Having had unsatisfactory views of a distant Great Knot Calidris tenuirostris at the end of June 2000 he was looking again, as he had every year since, in all the suitable spots around Wexford, Ireland. As he scanned through large flocks of Black-tailed Godwits Limosa limosa and Common Redshanks Tringa totanus on Our Lady’s Island Lake he picked up a dowitcher. Although it was summer the bird was in winter plumage. He videoed it and could see that the identification was not straightforward. Then it took off and flew straight over his head, but it didn’t call. Frustrating because, as we have already shown, had it called that would have identified it. Killian returned home and that night, having gone through all the feather tracts by watching the video in detail, decided that it was a Short-billed Dowitcher and clarified the identification on the bird news services. He returned the next two days and on the second day it called, confirming his identification.

Now all he had to do was record it for us. He returned over and over again but the loudness of the redshanks drowned the dowitcher’s calls and the conversational calls of the godwits sounded surprisingly similar. It was very frustrating and despite repeatedly sitting with the birds, Telinga poised and tape running, it took until the 20th of July to get it recorded. Even then, his recording had to be edited to remove an annoyingly close redshank call in order to create a less confusing recording (CD1-65). The Long-billed Dowitcher in CD1-66 was also a vagrant, one that spent two winters in the Netherlands. This bird was nearly always found in the company of a Common Redshank. Fortunately, this time it was on its own, and it was recorded before flying back to join its friend. It would have been easier to use the recordings we have from North America but it’s far more exciting to hear genuine vagrants.

Cracking the code 

The best way to take bird sound identification beyond ‘the call, the song’ thinking is to start comparing broader repertoires. To get an idea of the repertoire of a given species, we normally start with BWP, the only publication that really gives enough detail apart from the German Stimmen der Vögel Europas by Bergmann & Helb (1982). Most closely related species will have parallel call or song types for a range of behaviours, because their calls will have evolved from the repertoire of a shared ancestor, and these parallels are called homologies. Certain sounds may have evolved little from ancestral calls, while other sounds differ strongly between closely related species.

An extensive account of parallel repertoires and how bird sounds have evolved within a family is to be found in Bird sounds and their meaning by Rosemary Jellis (1977), one of the best introductions to the evolution of sounds. Jellis summarised and developed the work of many researchers who were “cracking the code of bird sound communication”. The book draws most of its examples from the tits Paridae and describes the sounds common to the whole family. It then shows how each tit species has evolved some of these common sounds into its own calls and song patterns, using subtle modifications. 

It has always annoyed me that Great Tit Parus major seems to be the only bird most birders credit with an extensive repertoire. Maybe it’s because Great Tit can make sounds that are mistaken for Common Chaffinch or other tits, and many birders have experienced the embarrassment of misidentifying these common species.

It’s interesting to explore how tits vary their scolding chick-a-dee sounds to create different calls. These are very useful sounds and slight variations are thought to tell one flock from another (Mammen & Nowicki 1981) and where the bird is in relation to others in the flock (Hailman & Ficken 1996). In North America the tits are called ‘chickadees’ and in a study it was found that when Black-capped Chickadees used calls of this type to mob predators, the length and number of dee notes (there is seldom just one) vary according to how dangerous the predator is (Templeton et al 2005). Willow Tit P montanus used to be considered part of the same superspecies, and its chick-a-dee calls vary the number and length of the dee notes along similar lines. The chick-a-dee sounds are also used as simple songs, or in more varied singing bouts. Listen to the recordings of this type of calls from four different species. In the sonagrams, the chick-a notes look like extreme variations of a similarly shaped sound. The dee-dee-dees are all broadband sounds, but with different lengths and delivered at different speeds. These recordings show how the tits have each evolved their own versions of these sounds.

In Dorset, birders have started to worry about the breeding population of Willow Tit. At the same time there is a greater understanding of the identification criteria for both Willow and Marsh Tits P palustris; features like wing panels, shape of bib, and overall size no longer stand (Scott 1999). So now it is not certain whether there has been a severe reduction in the breeding population of Willow Tit or there has been a change in the way people separate them. In other words, whether many sightings in the past concerned misidentified Marsh Tits. Fortunately, the differences in the chick-a-dee sounds are diagnostic. 

Willow Tit’s chick-a-dees are the most noticeable calls in its repertoire, and tend to be just a robust, consistently slow, series of long dees. Unlike Marsh Tit they don’t always start with their high-pitched, short chick-a (CD1-67). The equivalent calls of Marsh Tit do start with a sharper, whiplash chick-a, and the repeated dee notes are much shorter. They are also delivered more rapidly and while this varies with the degree of the bird’s excitement, the speed is typically twice that of Willow. Compare the dee notes in the sonagrams; those of Willow are like solid tower blocks with regular gaps in between. The weaker ‘Eiffel’ towers of Marsh show that its dee notes are less harsh, and the higher pitch of the fundamental makes the timbre sound less aggressive. This combined with the rapid delivery gives the impression of Marsh being a more excitable bird, although the sound is less distinctive (CD1-68). 

Arnoud recorded the alarmed Great Tit in the next recording while waiting for a Eurasian Pygmy Owl Glaucidium passerinum to call in southern Germany (CD1-69). At 7 kHz, the chick-a is very high pitched and will be missed by some of us, while the dee notes have a low-pitched fundamental and a rapid delivery. The European Blue Tit Cyanistes caeruleus in CD1-70 has a much faster variant of the same sound, in this case being used to scold the Eurasian Jay Garrulus glandarius in the background. Its chick-a sounds can be varied in all regards. Sometimes they are repeated several times or dropped for several phrases, and while high pitched, they are easy to hear. In the sonagrams, the dee-dee-dees vary in thickness (length) and subtly rise in pitch, dropping from the high pitched chick-a and rising at the end. This seems very different from the simple dee-dee-dees of Great, Willow and Marsh Tit. Now as you cast your eye over all the sonagrams, you should be able to see the relationships and the variations quite clearly. 

Parallel lines 

Closely related pairs of birds have parallel repertoires, a series of equivalent sounds that can be used in similar circumstances. Working through these enables you to decide which comparisons are the easiest for identification and where a sound is diagnostic. To illustrate this, we chose to take a close look at the sounds of adult Arctic Tern S paradisaea and Common Tern S hirundo. Getting the recordings needed for this section involved a lot of travelling. While I have hundreds of Common Terns breeding nearby on Brownsea Island, Dorset, I don’t have Arctic Terns, and anyway for recording purposes it is very noisy. Near Killian’s home in Wexford, there are sites for both species. Magnus and I visited Killian a number of times and we should have been able to make all the recordings for this comparison. Unfortunately, when you visit a mixed tern colony, the sounds seem to be an impenetrable chaos of endless variations and,  in the best month of August, Killian lures you away to look for vagrants at Tacumshin. Then when you get home you realise that you’re still missing some recordings. So far, The Sound Approach has visited 42 countries and this eight track section involved no less than five of them. We used two recordings from Wexford, while the others are from Kazachstan, Norway (the northern reaches of Finnmark), Spain, and finally birds recorded in Orkney and Shetland, Scotland. The five of us have travelled almost as far as an old Arctic Tern.

Only when I take the time to listen carefully at home to these recordings do the sounds become a reliable means of separating these very similar-looking species. The three most useful vocalisations for separating adult Arctic and Common Terns are long calls, advertising calls and kip calls. We’ve also shown scolding gyarrrr calls to illustrate a similar shared sound.

The simplest way to tell the two species apart vocally is the striking difference between the long calls. The long call of Common Tern (CD1-71) was recorded over Our Lady’s Island Lake in Wexford. It’s a rapid series of kearrip notes that speed up towards the end. Imagine skimming a stone across the lake and as the skips become closer and closer and lower and lower that’s how the sound speeds up with each note getting a little lower. The Arctic Tern’s long call (CD1-72) always consists of a repeating pattern containing a couple of staccato notes followed by a longer one: da-da-daaa, like a morse code SOS being tapped out with increasing urgency by a Norwegian resistance worker in a second world war movie. 

A relaxed and confident Arctic or Common Tern, flying from A to B, or just fishing, will often give calls that BWP have named advertising calls at fairly regular intervals, typically at a rate of about two every three seconds. Research has shown that these calls identify the individual, enabling a fledgling tern to recognise its own parent bringing fish. For us, these calls are not quite so easy to separate (CD1-73 and 74). The key differences between these two species’ sounds are pitch, speed of delivery, and length of note. Taking the geographical mnemonic to its furthest point: our Spanish Common Tern takes it a little slower, lower and longer, more mellow comparatively, keearr.., while the Norwegian Arctic Tern is sharper at first but then harsher, shorter and more hurried irrr.

When terns fly up and around you, scolding but not actually attacking, fewer advertising calls will be heard, and a more deliberate and harsher gyarrrrr, similar in both species, will become the dominant sound. This call is a variation on the advertising call and so again is slightly lower pitched in Common Tern (CD1-75) than in Arctic Tern (CD1-76), but this is only really noticeable in the fundamental frequency of the start of the call. Although not very helpful for species identification, it helps to distinguish this generic call from the more helpful advertising calls. 

Kip calls (CD1-77 and 78) can be heard in a variety of excited situations, typically involving retreat or departure. The pitch is again the key for identification here with the Arctic Tern higher pitched than the Common Tern.

Different species can share very similar sounds

If a particular sound is exactly right for a basic need, several species may use it even if they are not closely related. One of the first examples of this was described by Marler (1959) when he showed that many passerines (eg, Dunnock Prunella modularis, Common Blackbird, Common Chaffinch) use a high-pitched seep call in the presence of certain predators, especially hawks. He explained that the high frequency and drawn out nature of the call, as well as the soft start and end, give the call ventriloquial qualities. All three features of the sound make it very difficult for a predator to tell the location of the caller. At the same time, all the small birds in the vicinity are warned about the threat, and can take evasive action or just keep perfectly still. Listen to the examples of seep calls from unrelated songbirds: a Bohemian Waxwing Bombycilla garrulus in Iceland that has seen a Merlin F columbarius (CD1-79a), and a Ring Ouzel T torquatus in Turkey that can see and hear the Long-legged Buzzard Buteo rufinus calling very distantly in the background (CD1-79b). As an example of seep calls from a very common bird, listen to a European Robin Erithacus rubecula recorded in northern Scotland (CD1-79c). Often a cat, or in this case the recordist, is enough to stimulate calls of this kind.

One time while I was sitting in Białowieza forest in Poland listening for woodpeckers, I became aware of Common Chaffinches scolding. When I looked, I found a pair attacking a Great Spotted Woodpecker, which can be a predator of eggs and nestlings (CD1-80a). Calls used aggressively as threats can be very similar between unrelated species, and nearly always are used at close range, loud and harsh-sounding in most birds. To an opponent, the sound is literally ‘in your face’; by using as many frequencies as possible a bird makes itself seem more powerful, more present, like a cat making its fur stand on end to look bigger. One aspect of making recordings is that you can make comments on the tape as a note for later. The disadvantage to this is all your mistakes are there for posterity. When I recorded the Corn Buntings Emberiza calandra on Mallorca in CD1-80b, I can hear myself saying; “male came to chat with another male that had been singing”. On reflection, male birds during the breeding season don’t usually chat: they argue, and sure enough the harsh calls from four seconds into the recording sound almost identical to the calls we heard in the chaffinch recording. Listen to another example, recorded at a feeder in my back garden. In this case, European Goldfinches are threatening each other, as they squabble over food (CD1- 80c). The goldfinches deliver the calls with the same noisy use of a broad range of frequencies, but with a different more rattling rhythm. 

Calls of this type are not limited to songbirds, and the gyarrrrr scolding calls of Arctic and Common Terns fit this category (CD1-75 and 76). You can even hear something similar in some wildfowl. I particularly enjoyed making the next recording in Norwegian Lapland, sitting sheltered from the wind by a large rock facing a sunny tundra pool with several broods of Common Eider Somateria mollissima, three female Red-necked Phalaropes Phalaropus lobatus, some Ruff Calidris pugnax and a pair of copulating Parasitic Jaegers Stercorarius parasiticus. Just when the eider were on the point of leaving the water, a Hooded Crow C cornix showed up, possibly interested in their young. The females barked at it angrily and all the eider, young and old, got back in the water (CD1-81). As a sonagram shows, the calls the eider used to threaten the crow are like a much deeper version of the other threat calls you already heard; and the noisy structure, maximising the range of frequencies used, is very similar.