An Ornithologist Finds That Nature's Feathered Musicians Aren't Just Whistling Dixie
Is there a difference between birdcalls and birdsongs?
Yes. With practically any bird you can catalog a repertoire of the sounds it makes. Birdcalls are typically short units of sound—chirps, clicks and so on—used for announcing danger or food or courtship. The clucking of a chicken is a birdcall.
And what are birdsongs?
These form another pattern of sound, longer and more melodic and structured. When a cockerel, that is, a rooster, crows, that's his version of a song. It's a territorial proclamation speaking primarily to rival males. It's also attractive to hens, a signal to stay around or, if unattached, that here's the possibility of a potential mate. Those are basically the two functions that dominate birdsongs in general.
Do all birds sing?
No, but they all make sounds of one kind or another. Of some 9,000 species worldwide, about 4,000 are classified as songbirds. Usually only the male sings, although there are exceptions such as the red-winged blackbird. Some tropical wrens, for instance, often sing highly structured duets, like part-songs; the male for his part sings one role while the female for her part sings another.
Why are there so few chanteuses in the songbird world?
In male songbirds the brain nuclei controlling song are very large and extraordinarily well defined. But if you look at the brain of a female white-crowned sparrow, for instance, or female canary, you see the same nuclei, but they're smaller. The females of both these species have the latent ability to sing, and you can induce them to do so by administering male sex hormones. However, in the female zebra finch you will hardly see song-control nuclei at all. Biologists once assumed that in the higher vertebrates the brains of males and females were virtually identical in structure. This is definitely not so in songbirds, and it was the discovery in 1976 of sex differences in songbird brains that led to a revision of a whole position. There are even strong suggestions of sex differences in parts of the human brain, although scientists argue about their nature and significance.
Which songbird species is the all-time musical virtuoso?
After weeks of analyzing sound charts, a colleague of mine established that a single brown thrasher living in the field near here sang more than 2,000 different songs. As far as I know, that's a record. There are people now working on mockingbirds that may topple that record. A more normal range for songbird repertoires runs between five and 20 selections, though quite a few species have around 100.
Why such awesome repertoires?
Sexual prowess is one possibility. Males may actually have a hierarchy among themselves. When you have two male wrens countersinging, for example, one always leads the other. Maybe the one with the larger selection is more vigorous, maybe he's older. We have some evidence that some female birds, if given a choice, will favor the older male.
Any other reason?
After a pair-bond has been established, there's often a long interval before the female actually builds a nest and is ready to copulate and lay eggs. In some species there's evidence that the complexity of song is important in ensuring the female's progress through all those stages as soon as possible. Mockingbirds sing almost continuously, as if to barrage the female with song. Of course, if you sing the same thing all the time, the female gets bored, and it looks as though the diversity in some birdsongs is simply to cut down marital boredom.
Are birdsongs instinctive or learned?
The rules are different in each case. A cockerel will crow in perfectly normal fashion whether or not he has ever heard other roosters. But among songbirds proper the song is always learned, in the sense that if you take a young male and rear him out of hearing of its own kind, then, although he will sing, the structure of his song will be abnormal. The notes tend to be less melodic, the repertoires much smaller. So in songbirds you get this interplay between genes and environment.
Can one species be taught to imitate another?
Starlings and mockingbirds mimic other birds quite normally in their natural song. It's a bit of a mystery why they do that. Mockingbirds are very aggressive, and they attack not only other mockingbirds but other species as well. There's been a little bit of evidence they might defend their space more effectively by using sounds of other species, but the evidence is not very robust. A better example is the lyrebird of Australia, which imitates not only other birds but also mammals, mechanical sounds, even the twanging of barbed-wire fences. But the vast majority of songbirds learn selectively. They have the natural ability to filter the sounds they hear when they are very young to favor the songs of their own species. For example, a young swamp sparrow is innately able to reject the songs of even its close relative, the song sparrow.
What about parrots or mynah birds?
So far as we know, they never mimic other species in nature. It's only when you withdraw a parrot from its natural community, and it's looking around desperately for something else to establish a relationship with, that it begins to attend to your speech and imitate human sounds. In nature they use their ability to mimic entirely for learning dialects within their own repertoires.
By dialects, do you mean that the song of one species varies over different geographic areas?
Yes, and it is often as distinct as in human speech, as much a variation as that between, say, British and Southern accents. If you have a good ear, you can tell where you are in California simply by listening to the white-crowned sparrows. Each dialect area involves no more than a few hundred birds, with very well defined boundaries between dialects. You can bring a very young bird into a laboratory and train it to sing in a "wrong" dialect to show that this is all learned behavior. This ability to learn song is a rare thing among animals, and there are parallels between a bird learning to sing and a child learning to speak.
What are the parallels?
In young songbirds the equivalent to an infant's babbling is what we term subsong. The bird has to learn how to control its voice by ear, a very refined and special skill. In the transition from subsong to full song you get what you would expect in a bird that's honing its skill, consulting memory for the kind of song patterns it heard when it was young. That's exactly what happens when a child begins to speak, with the babbling serving to help shape its individual note structures, its own culture and finally to develop the local dialect. Subsong is the only analogue we have in animals for babbling, so it's a unique opportunity to study how an organism that must learn to sing actually goes through the process of generating sounds on the basis of what it has heard.
Do songbirds learn continuously through life?
The sparrows we have worked with didn't learn any new songs after the first month and a half of life. But a male canary replenishes his songs with a new repertoire every breeding season, and, so far as we know, he will do that until he dies. With canaries, though, the song repertoire is finite in any one breeding season. With something like the brown thrasher, you even have to face the possibility that it continues to add songs throughout the breeding season, so that there's no period when the repertoire is set over even a couple of months' time.
Is it possible that, like people, songbirds sing just because they feel like it?
Why do I sing in the shower? Presumably because I feel good. And, yes, that's probably a relevant answer for birds too.