Daniel Crawford’s “A Song of Our Warming Planet” is striking for its rising pitch as the Earth warms, but it reveals something deeper about nature and what we perceive as “music.” (More)
As addisnana shared here on Wednesday, University of Minnesota cellist Daniel Crawford worked with geography professor Scott St. George to convert annual mean temperature data since 1880 into music. The concept of data sonification is not new. We easily recognize the rapid clicks of a Geiger counter or the rising squeal of a metal detector. But presenting temperature data as music was a novel idea, as Dr. St. George told Ensia:
Data visualizations are effective for some people, but they aren’t the best way to reach everyone. Instead of giving people something to look at, Dan’s performance gives them something they can feel.
And you can feel the change in Crawford’s composition:
The first change, the rising pitch as the Earth warms through the 20th century, is obvious. But listen closely and you’ll hear another, more subtle change.
From 1880 to the 1940s – 1:36 to about 2:30 in the video – “A Song of Our Warming Earth” sounds very musical. The melody doesn’t follow any familiar chord progression, but you might easily mistake it for a work from Béla Bartók’s experimental period. But by the 1970s – from 2:45 to the end of the song – it sounds less like music and more like … just noise.
The difference has to do with what “music” is and, specifically, what a musical melody imitates in nature. Consider this passage from The Science of Fractal Images, edited by Heinz-Otto Peitgen and Dietmar Saupe:
With the exception of very modern composers like Stockhausen, Jolet, and Carter (where the melody fluctuations approach white noise at low frequencies), all types of music share this [inverse frequency]-noise base. Such a view of melody fluctuations emphasizes the common element in music and suggests an answer to a question that has long troubled philosophers. In the words of Plato, “For when there are no words (accompanying music), it is very difficult to recognize the meaning of the harmony and rhythm, or to see that any worthy object is imitated by them.” Greek philosophers generally agreed on the imitative nature of the arts. It seemed obvious that painting, sculpture, or drama imitated nature. But what does music imitate? The measurements suggest that music is imitating the characteristic way our world changes in time. [Emphasis added]
It turns out that our ears are keyed to the same not-quite-random patterns we might see in a coastline, a treeline, or the edge of a cloud. That not-quite-randomness can be measured as a fractal dimension, and we’re used to seeing a fairly narrow range of fractal dimensions in nature. For example, the coastlines of Ireland, Great Britain, and Norway have fractal dimensions of 1.22, 1.25, and 1.55, respectively.
And that’s about the same range of fractal dimensions that generate melodies we recognize as “musical.” Lower dimensions are too predictable to hold our interest and sound “boring.” Higher dimensions, like a toddler randomly hitting keys on a piano, sound like “just noise.”
And that’s why the second half of Crawford’s composition sounds so “wrong.” It’s not simply that the pitches climb with the warming of the Earth. The early melody fits a pattern we’re used to in nature, and we hear it as “music.” But as the Earth warms and the pitch changes become more erratic – a predictable effect of a more energetic atmosphere – that “musical” Earth-song becomes … “just noise.”
You can hear that change without knowing the first thing about Brownian motion or fractal geometry. It feels wrong – more and more like “just noise” – and that’s what makes Crawford’s composition so effective.