Blue is a beautiful color, but its sound is simply irresistible. It is the song of the unhappy and the depressed. It is a sound that touches people. It was also the sound of a little songbird, the Serinus Canaria Domestica, a sound that so moved me, I was led on a voyage of discovery into the world of birdsong. The Serinus Canaria Domestica is the man-made descendant of the Wild Canary, a finch originally from the Canary Islands, which nowadays exists in many different breeds. This essay deals with the cultivation of the song-bred canary and imagines how its story might lend inspiration to the sound design of electric cars.


Sounds ‘exist in time and over space’ [1]. You can hear a sound without having to face the source that produces it; you only have to be listening or recording at the right time. If you want to see an object, however, you have to be facing it. And, in most cases, you can re-view the object at different moments and for longer periods; visual objects therefore ‘exist in space and over time’.

When you are a small bird living in dense foliage, leaves prevent effective visual communication. This makes sound an excellent alternative for warning or impressing your mates, or for marking out your territory. The volatile character of sound, however, makes its evolutionary development difficult to trace, whether it be birdsong or vocal communication in animals in general. We know from visual fossil inspection, for example, that there was a close relationship between dinosaurs and birds [2]. At the same time, though we have a sense of what dinosaurs looked like, we can only imagine their vocal expressions today.

In On the Origin of Species, Darwin explains how adapting to changing conditions in the natural environment results in survival for some living organisms and extinction for others. Biologists have discovered that this principle of ‘natural selection’ not only causes species to develop subspecies with very different characteristics that are determined by heredity, but also lies at the basis of the origin of new species [3]. The Domesticated Canary is a subspecies of the Wild Canary and contains a wide variety of breeds that have not been scientifically classified. The origin of these (new) breeds is a result of ‘human selection.’ But what exactly does this ‘human selection’ principle entail? And can this principle inspire, or maybe even guide, the sound design of next nature? Before focusing on what comes ‘Next’, I will briefly review the ‘Current’ ground of Nature’s infinite design space as cultivated by human breeders of the species: Canaria.

Nowadays, three main groups of domesticated canaries can be distinguished: posture, color, and song canaries. The various breeds within these groups show a wide variety of different shapes (small, big, curved, bowed, curly-feathered, crested, and more…) and colors (green, yellow, red, brown, white, orange, gray, and more…, though no blue!). The song canary group comprises different breeds with clearly distinguishable songs. Unfortunately, the richness and uniqueness of these different songs cannot be captured in words; a ‘sound’ description would take pages! For now, I will introduce two of the most familiar breeds of this group: the Harz Roller (a.k.a. the German Roller) and the Waterslager (a.k.a. the Malinois).

The song of the Harz Roller canary was cultivated in the Harz Mountains in Germany, whereas the Waterslager originates in Belgium. The melancholy song of the Harz Roller is characterized by relatively slow, nostalgic, soft accents as compared to the jubilant song of the Waterslager, which has a more animated rhythm with sound segments (tours) that are more individually distinct [4]. Although these song-bred canaries sound very different from each other, the Wild Canary is their shared ancestor. What selection principles were involved in the breeding of these distinct songbirds?

To answer this question, we will assume the vantage points of the range of actors involved in the evolutionary process. The male bird is the lead character—he’s on lead vocals. He’s the only one that sings; female canaries, and female birds in general, do not sing. And he had better sing well (!), to impress the female canary, create a bond, and bring about a successful mating. In our case, however, the act of singing clearly goes beyond the mating function: the male bird not only has to please the female bird, but the human breeder as well. Unbeknownst to the male bird, it is ultimately the breeder who decides for or against the composition of a possible breeding pair based on the song qualities of the male bird.

However, there is an important difference: the breeder’s (= human) selection criteria predominantly relate to the aesthetic qualities of birdsong, whereas the functional qualities of the song of the male bird seem to dominate natural selection principles. Female birds judge a male bird’s physical fitness for reproduction on his vocal performance. Yet any person who has ever listened to the varied, beautifully nuanced, and apparently improvised phrases performed by a solitary songbird with no other birds in its direct vicinity might seriously wonder whether reproduction is the only intrinsic motivation for birds to sing [5]. The third principal actor is the female bird. She not only has to be susceptible to the male’s singing courtship behavior, but she should also supply a good genetic blueprint for nesting behavior, as this is what determines the actual offspring produced in any generation.

In the case of the Harz Roller and Waterslager breeds, breeders’ opinions about what made the perfect canary song differed sharply. The Harz Roller breeders preferred low, smooth, rolled sounds above shrill, noisy sounds, leading to the calm, melodic song of the Harz Roller as it is known today. For the breeders of the Waterslager canary, on the other hand, the song of the Nightingale was the model to emulate. This led to the interrupted, boiling and rolling water beats and metallic tone qualities that characterize today’s Waterslager song. Already in the nineteenth century, breeders organized clubs to share knowledge and to hold song contests. Standards of song quality were first established within these clubs, and eventually led to worldwide standards describing the various song tours and their ideal qualities.

Scientists have recently discovered [6] that these canary breeds differ with respect to hearing sensitivity for high-frequency sounds. Waterslager canaries show impaired hearing in the frequency range in which their vocalizations contain the most energy. In other words, in order to contact a ‘hard-of-hearing’ female, a male Waterslager has to produce louder sounds. This finding demonstrates that the non-singing female birds have an equally important role in the evolutionary emergence of new song-bred canaries. At this point, the case of the Serinus Canaria Domestica has been introduced in sufficient detail to address the main question of this essay: how can the cultivation of traditional nature inspire Next Nature’s sound design? For this purpose, and as a hypothetical example, I will consider a challenge currently faced by car manufacturers—sound design for electric cars.

Car manufacturers have known for quite some time that the sounds their cars produce need to be explicitly designed. While the functional quality of car sounds guarantees skilled and safe driving, their subjective qualities are crucial to the driver’s overall experience, as well as the car company’s brand image. Consider the subjective associations of a car door slamming or an accelerating car engine. A car that does not produce (the right) sounds has the same effect on the driver’s experience as a silent movie played on a full-blown, state-of-the-art home theatre system.

By mapping the lead characters of the song-canary case directly onto the stakeholders involved in the sound design of future electric cars, some intriguing new interactions immediately pop out. The car (male bird) produces the sounds that will impress and seduce its future owner (female bird) into purchasing.

The sound designer or car manufacturer (breeder) decides which car and corresponding sound set best matches a particular customer segment. This may sound like common practice, but songbirds and their ‘designers’ do things differently. First of all, their songs are dynamic and adapt gradually to the changing environment. Moreover, as we have seen, cultivated birdsong goes beyond the functional, and the aesthetics of expression are at the heart of its being. For future electric cars, this could mean that the basic ‘brand specific’ sound synthesis algorithms and the type of sounds they are able to generate will still be defined by the car manufacturer, but that individual cars may be able to learn sounds and adapt them to their own environments and driver preferences. In this scenario, a Ferrari will always sound like a Ferrari, but a Ferrari from the countryside will easily betray its origin by sounding completely different than an urban-raised Ferrari. More ‘open’ futuristic scenarios would allow any car to disguise itself as a Ferrari sound-alike [7], or even audiomorph into a Batmobile destined to break the sound barrier.

Other adaptive schemes could breed ‘cars with personalities.’ A future car, for example, could adapt its sound to its owner’s driving style, or sonically radiate the driver’s personality traits. Such sonifications would enable drivers and their environments to become aware of behaviors which, if desirable, could boost self-esteem or, in the case of unwanted behaviors, could motivate for behavioral change. And what about car-driver units synchronizing their sounds to those of other car-driver units, much like cicadas sometimes synchronize their songs, or as song-canaries have been trained to sing in pairs or in groups of four? Such emergent phenomena could create positive feelings of being connected and, at the same time, improve traffic flow.

Many more scenarios could be envisioned, but the most important challenge remains to create the right conditions for an ecosystem to emerge in which all stakeholders (car manufacturers, intelligent cars (!) and car drivers) will be able to freely explore the opportunities offered by sound. As we have learned from the case of the song-bred canary, these explorations need to be determined by interactions between the various stakeholders. The conditions for interaction need to be defined properly in order for this kind of evolution to thrive—one in which brand-specific sound sets simultaneously reflect the personal preferences shaping the driver-car relationship. Only then will there be a chance that one day, at daybreak, I will be moved again, this time by the sad song of a lonely abandoned car, subtly standing out from the peaceful dawn chorus in my backyard.


[1] Gaver, W.W. (1989). The Sonic Finder: An interface that uses auditory icons. Human-Computer Interaction, 4 (1), 67-94, 1989.

[2] Ruben, J. (20 10). Paleobiology and the origins of avian flight. Proceedings of the National Academy of Sciences, 107, 2733-2734, 2010. 2010; or for a popular summary see:

[3] Orr, H.A. (2009). Testing Natural Selection. Scientific American 300, 30-37, 2009.

[4] World Confederation of Ornithology: Song Standard of the Waterslager/Malinois Canary. retrieved on 02-09-2010.

[5] Rothenberg, D. (2005). Why Birds Sing – One Man’s Quest to Solve an Everyday Mystery. Penguin, Allen Lane, Great Britain. 978-0-713-99829-6; also see accompanying website: retrieved on 02-09-2010.

[6] Okanoya, K., Dooling, R.J. and Downing, J.D. (1990). Hearing and vocalizations in hybrid Waterslager-Roller canaries (Serinus canarius). Hearing Research, 46, 271-276.

[7] Hukar Ozyasar (2010). How to make my car sound like a Ferrari. retrieved 21-10-2010.

Enjoying this story? Show it to us!


Share your thoughts and join the technology debate!

Be the first to comment

More like this