“They charm the females by instrumental music of the most varied kinds”
And thus, Charles Darwin adapted the phrase “Instrumental Music,” previously used to mean humans with instruments making music, to name one of the most important “secondary sexual characters … diversified and conspicuous in birds” which, added to “all sorts of combs, wattles, protuberances, horns, air-distended sacs, topknots, naked shafts, plumes and lengthened feathers gracefully springing from all parts of the body” mediate avian sexual selection1.
Bird “instrumental music” is bird song. Bird song is defined in various ways, but I prefer the functional definition used by evolutionary biologists. A bird song is a communication having to do with sexual selection. The modal bird song is made by a male to signal his quality to females who are busy choosing among a number of possible nest mates. That same song may be thought of as a territorial marker, but the territorial nature of bird song is probably more of an atavism than a demonstrated fact. If keeping other males away is a function of song, it is probably a secondary one. The song is better thought of as a quality of the territory and the male, and that quality indicator may well have ‘meaning’ to both male and female birds.
The modal bird song is a song, but not all ‘songs’ are song-like, in the sense that we expect song to come out of the ‘mouth’ (bill) and to consist of modulated air currents (like human voice). Just as human language is not always spoken (it can be signed, for instance, or in some cases, transmitted with a certain look) “bird song” may be generated by beating of feathers, clattering of a bill, or some other noise-making somatic activity.
One of the most spectacular examples of communication among birds that is all about mating but involves no song is the bower and gifts assembled by the male bower bird, where just like in humans, a nice place to hang out, a pile of cool but useless presents, and excellent dancing often leads to a second date.
Just as importantly, not all bird sounds are bird songs. One of the (undesirable, in my view) definitions of bird song is that it must be complex. Thus, the cooing sound of a pigeon is not a bird song. But it is, if it is what Male pigeons do in relation to mating and sexual selection. But other sounds, like the hard to localize tsick tsick of nuthatches or chickadees which we may refer to as “contact calls” are not songs. They are modulated air streams coming out of the beak, they are sounds, they may have a role in the percussion section of the “instrumental music” of the avian cacophony, and they may even be social communication at that, but no matter how sexy a nuthatch’s tsick tsick may be to a birder who has not seen a nuthatch in a while, it is not going to get the bird a date or even a glance (well, not that kind of glance).
So defined, bird songs are the sounds, often but not always loud and musical, that (mostly) male birds make as part of that species’ mating system. This kind of vocalization may have evolved a few different times, though I for one am not going out on a limb to explain what some call “songs” in hummingbirds or parrots. If we think of bird songs as whatever sounds male birds make for mating purposes, then there are far more parallel examples among the birds. If we require that bird songs sound like the sounds that “song birds” make, then, well, this behavior evolved once, in the song birds (duh). Sticking strictly with the usually melodious and mating related form of the behavior, as is found in the Passeriformes is a good bet and allows us to talk about a well studied and understood phenomenon.
These bird songs have two aspects, and I oversimplify but not excessively so: A genetically inherited template, usually specific to the song-producing sex, and a learned song that consists of the template shaped by experience as a developing bird hears the adult version of its own song. The templates can be discovered by raising birds in the absence of adult song, and they often sound very little like the normal final product.
These are sonograms representing the song of four Zebra Finches. The top two are individuals with normal song. The bottom two are the songs of individuals raised in isolation, representing only their “built-in” template unaffected by the experience of hearing normal adults.
In these song birds, we often see dialect differences between regions. If a hatchling is raised in the region in which a dialect other than that of its parents is used, it will develop its genetically determined and undifferentiated template into the dialect used in the area to which it is moved, just like a French infant conceived of French-speaking humans in Paris but raised by Americans in Dallas, Texas will speak English.
The most important thing about this is the following: The parts of the bird song that determine the Darwinian success of a bird … via its ability to find a mate or to do whatever else we find the bird is doing with its song … is not determined mainly or even measurably by the template for that song. It is, rather, determined by its correct learning of the song as it is sung locally by members of its species, plus other non-genetic factors we will not investigate today.
This is important because when we think about evolution, we often (because we are humans, and humans are not all that smart) immediately simplify the world into one in which organisms carry (variant copies of) genes that they (may or may not) pass on to offspring and that determine that organism’s behavior. In this simplistic view, anything that varies that is not genetic is noise. But this can not possibly describe bird song. In song birds, the song is learned, and it is learned in a way that is determined entirely by local conditions which clearly change across space from region to region, as a dialect of a widely spoken human language may vary from place to place.
Putting it a slightly different way, the template from which a bird builds its song is passed on genetically, and the song itself is passed on culturally, or if you prefer a different term, “extrasomatically” or “non-genetically” or whatever you like. But not genetically.
This parallel transmission is certainly “Darwinian” and evolutionary in both tracks. A gene that has a mutation causing the protein it expresses to be ‘broken’ in a way that messes up the bird’s template will find itself in a mute male bird and thus not be passed on. A variant of a gene that produces a template that has some sort of advantage might be passed on to more offspring. But the same is true of the song itself. Imagine starting with two sets of Zebra Finch hatchlings. One group is exposed to normal zebra finch song for a given area, the other to a fake bird song that is sufficiently like a zebra finch to cause the birds to learn a “song” but one that would be ineffective in its social role for wild zebra finches. Now, release the finches. Which ones will contribute both their genes and their song to future generations? Unless the fake song you used in this experiment was, accidentally, an extraordinary song, the basis for the next avian idol, the birds in the second group and their ‘song’ would be evolutionary dead ends. Their genomes would be an evolutionary dead end AND their song would be an evolutionary dead end.
A variant song that is hard to learn, inaudible to other birds, or that causes some sort of inappropriate response or confusion will undergo Darwinian selection just as a variant gene that causes a bird to have an ineffective template will undergo Darwinian selection. Minimally, bird songs are selected to be something a bird can generate and something another bird can hear, against the background of ambient sounds, at the distances at which these birds are communicating, and in the birds’ local cultural context.
A recent series of demonstrations has been published of how plastic bird song is, and how the learned part of this parallel process of transmission can be adapted, in a Darwinian sense, just as a population of genes might be adapted, to changing conditions. Birds living in urban environments must sing against different background from their rural counterparts. Urban noise covers over sounds at certain, typical lower, frequencies, and may also drown out other sounds in general. The following studies show that there may be an effect on the bird songs (not the genes) in urban environments:
… bird songs are modified in different ways to deal with urban noise and promote signal transmission through noisy environments. Urban noise is composed of low frequencies, thus the observation that songs have a higher minimum frequency in noisy places suggests this is a way of avoiding noise masking. … we experimentally test if house finches (Carpodacus mexicanus) can modulate the minimum frequency of their songs in response to different noise levels. … We conclude that house finches modify their songs in several ways in response to urban noise, thus providing evidence of a short-term acoustic adaptation.2
… We compared the songs of blackbirds, Turdus merula, from the city centre of Vienna and the Vienna Woods and found that forest birds sang at lower frequencies and with longer intervals between songs. This difference in song pitch might reflect an adaptation to urban ambient noise. However, the song divergence could also be the result of more intense vocal interaction in the more densely populated city areas or a side-effect of physiological adaptation to urban habitats. 3
We studied three adjacent dialects of white-crowned sparrow songs over a 30-year time span. Urban noise, which is louder at low frequencies, increased during our study period and therefore should have created a selection pressure for songs with higher frequencies. … We suggest that one mechanism that influences how dialects, and cultural traits in general, are selected and transmitted from one generation to the next is the dialect’s ability to be effectively communicated in the local environment.4
I’m fairly certain that if a similarly structured set of findings were presented for humans, there would be many who would jump in and insist that the change is genetic. Humans have fetishized their own genetic evolution to the point that if something is not considered genetic, then it may as well not be considered biological at all. But people who study birds know better. And, it may simply be the case that our feathered friends demonstrate parallel evolution more clearly than do humans, in part because the learning of song is accompanied by overtly visible neurological changes in the birds, while learning, when it happens in humans, does not. In any event, the songs of song birds focused Darwin’s writing about the evolution of humans “and selection in relation to sex” in 1871, and it tends to focus our thinking these days about the evolution of language and other cognitive facilities in humans.
So next time you hear a human blathering on about the genetics of human behavior, ask them to listen for a minute to the birds chattering outside. Ask if he thinks bird song is a genetically determined behavior. And use that special look you’ve surely developed by now indicating that this is a trick question. One raised eyebrow can be worth a thousand tweets!
1Darwin, C. R. 1871. The descent of man, and selection in relation to sex. London: John Murray. Volume 2. 1st edition. The quoted parts are from page 414. Online here.
2Bermudez-Cuamatzin, E., Rios-Chelen, A., Gil, D., & Garcia, C. (2010). Experimental evidence for real-time song frequency shift in response to urban noise in a passerine bird Biology Letters, 7 (1), 36-38 DOI: 10.1098/rsbl.2010.0437
3Nemeth, Erwin and Henrik Brumm. 2009. Blackbirds sing higher-pitched songs in cities: adaptation to habitat acoustics or side-effect of urbanization? Animal Behaviour
Volume 78, Issue 3, September 2009, Pages 637-641.
4Luther, David and Luis Baptista. 2009. Urban noise and the cultural evolution of bird songs. Proc Biol Sci. 2010 February 7; 277(1680): 469–473.
The image of the sonograms is from Lipkind, Dina and Tchernichovski, Ofer. 2011. Quantification of developmental birdsong learning from the subsyllabic scale to cultural evolution. Published online before print March 21, 2011, doi: 10.1073/pnas.1012941108 PNAS March 21, 2011.