The world's first podcast turning data into music. Join Duncan Geere and Miriam Quick as they use data sonification to create a series of original musical compositions from data about climate change, inequality, beer, and more.
The European Union is a truly unique political entity, and an economic, diplomatic, and cultural superpower. Perhaps the EU’s greatest achievement is that all this has been done through democratic lawmaking - not through armed force.
That lawmaking results in a lot of paperwork, and it’s this glorious bureaucracy that we wanted to celebrate when we came up with the idea to do a Loud Numbers episode on the EU, in the same year that our home country, Britain, left.
In this sonification, we took a database of all the laws ever passed by the EU and its precursors - a whopping 142,036 of them, from 1952 to 2019 - and created a piece of classical music that reflects the amount of lawmaking going on over time.
We turned the data into a fugue - a musical form where two or more melodic lines, or ‘voices’, interweave around each other based on strict rules. Each line begins with a short, specific string of notes, known as a subject. This subject is then repeated in every subsequent voice as it comes in, often in different keys, or sometimes even turned upside down.
There’s only one encoding - the number of laws made each year, which is mapped to the number of voices. Two bars of music equals one year of data. When there are only a few laws being made that year, there are only one or two voices. When there are a lot of laws, there are a lot of melodies – up to eight at once. It’s a simple mapping, but the result is far from simple.
And the subject of the fugue? You’ll recognise it as the first part of the European Anthem - Beethoven’s “Ode to Joy”, as arranged by Wendy Carlos in the soundtrack of Kubrick’s “A Clockwork Orange”. Our choice of retro synth sounds is a nod to Carlos.
The track covers the years 1952-2019 inclusive.
There’s just a single data reference: the European Union’s EUR-Lex Eurovoc database: https://eur-lex.europa.eu/browse/eurovoc.html
Boom & Bust tells the story of the US economy since the late 1960s – through data sonification.
It’s a rollercoaster ride of growth and decline, told through a UK jungle track that maps the Amen break to the state of the US economy each quarter. One bar of track time represents one financial quarter in the data. When the drum loop plays forwards, the economy is going strong. But when the US economy starts cooling down and spinning backwards into recession, so does the music. Listen out for the airhorn when the US economy starts growing again.
Additionally Boom & Bust sonifies the value of the Dow Jones industrial average, which tracks the performance of 30 leading companies in the United States. It’s mapped to a gentle, wispy synthesizer, which rises in pitch as the Dow Jones rises.
The track also tracks the dramatic rise of inequality since the late 1960s. In 1968, when the data begins, the poorest half of the US population took home about 20% of national income, while the richest 1% laid claim to about a tenth. Today, those proportions are reversed. The volume of the ‘badman’ sample represents the share of national income going to the richest 1%.
Samples of politicians and other notable figures are scattered throughout the track, telling the story of what’s happening at approximately that point in time. Finally, the bass drops in years when there was a presidential election.
The US economy is like a party: not everyone’s invited.
The track covers the time period from Q1 1968 to Q1 2020 inclusive.
Recession data was provided by the Federal Reserve Bank of St. Louis: https://fred.stlouisfed.org/series/JHDUSRGDPBR
Quarterly historical figures for the Dow Jones Industrial Average come from Stooq: https://stooq.com/q/d/?s=%5Edji&c=0&d1=19670101&d2=20200903&i=q
Can you hear a taste? We think you can. Tasting Notes is a data-driven musical representation of the taste of beer, based on scores compiled from beer expert Malin Derwinger. For this data sonification, we asked Malin to look at, smell and taste ten different beers and log her findings using a unique beer scoring system she has developed. Then we turned those taste scores into ten short pieces of music that mirror the sensory experience of cracking open a bottle and taking a swig.
Taste is complex, and so is this sonification. Each of the ten beers is based on ten different scores that reflect its aroma, mouthfeel, taste and appearance. Aroma is made up of three components: malt, hops and fermentation. Mouthfeel is made of two: body and carbonation. Taste is made of sweetness, alcohol, acidity and bitterness.
The louder the sound, the more pronounced the aroma or flavour component in that beer. And the character of each sound matches its taste. For example, carbonation, or fizziness, is represented by a rapid upwards sweep mimicking bubbles on the tongue. The fizzier the beer, the louder the sound.
Finally, we also included a tenth score for colour, based on how the beer looks. Lighter beers are shifted up in pitch, and darker beers are shifted down, with medium beers, well, in the middle. Each piece of music lasts between about ten seconds and a minute – beers with a stronger aftertaste linger longer.
We’ve brought these ten short pieces together into a track, accompanied by Malin telling us a little about each beer in turn. So grab a cold one from the fridge, sit back and enjoy!
Data was provided by Malin Derwinger, according to her standard beer-judging schema. Aroma was judged in the categories of intensity, malt, hops, fermentation and other. Taste was judged in the categories of intensity, carbonation, body, sweetness, acidity, alcohol, bitterness and aftertaste duration. Appearance was judged in the categories of foam, clarity and colour. We used all of these categories in our sonification except taste and aroma intensity, foam and clarity.
The full list of tasted beers comprises: Czech Pilsner, Alcohol-Free Lager, IPA (West Coast Style), Stout (Irish), Witbier, Fruit Sour, Gueuze, Rauchbier, Best Bitter, Strong Belgian Dark Ale, and Gose. We sonified scores from all these beers except the West Coast IPA (cut for length reasons!)
Every spring since 1916, the residents of Nenana, Alaska, have placed a tripod on the frozen Tanana river and placed bets on when the ice will melt, pulling it over. The measurement method has stayed the same over a century, making the competition records a valuable source of data for climatologists studying how the planet - and particularly the polar regions - are changing.
The Natural Lottery turns this climate data into a techno track. The higher the pitch of the chords in the track, the earlier the ice melted that year (using a 10-year moving average). These chords go up and down in pitch, but on the whole they get higher as the music progresses, showing the ice melting earlier and earlier as climate change in Nenana takes hold.
Two other data layers can be heard in the track. During the winter, the aurora borealis swirls through the skies of Alaska and its strength rises and falls in eleven-year sunspot cycles. These are sonified as an ethereal shimmer in the background, based on real data from the Royal Observatory of Belgium - the louder the sound, the more sunspots heard in a given month.
Then there’s CO2. In the background of the track, faint at first and louder and louder over time, you’ll hear a siren. The pitch of the siren represents carbon dioxide levels in the atmosphere, measured by the observatory at Mauna Loa in Hawaii.
They rise and fall each year as forests grow and die back in the northern hemisphere, which has more land. That’s why the pitch of the siren wobbles a little. But they also increase over time and rise to a worrying climax near the end of the track.
Finally, there are a whole lot of other musical elements that don’t represent any data. They’re just there to make the track sound good.
The track covers the time period from January 1917 to December 2020 inclusive.
Annual tripod data comes from the Nenana Book of Guesses, with data for the last few years gathered from news reports and appended manually.
Monthly sunspot data comes from the Royal Observatory of Belgium.
Monthly CO2 data for 1917-2014 comes from the Institute for Atmospheric and Climate Science (IAC) at the Eidgenössische Technische Hochschule in Zürich, Switzerland.
Monthly CO2 data for 2015-2020 comes from the US National Oceanic and Atmospheric Administration (NOAA) https://gml.noaa.gov/webdata/ccgg/trends/co2/co2_mm_gl.txt