Visualising the Tonal Characteristics of Sonic Youth
10th December 2018
Born out of New York’s late 1970’s avant-garde “No Wave” scene, Sonic Youth were an enigma to me as a teenager. I could play songs on the guitar that I liked by various bands, all except theirs. The problem being that they used a number of alternate guitar tunings which, in the pre-web days, were a complete mystery to me. Alternate tunings that meant it was very difficult, if not impossible to play their songs using the standard guitar tuning of E Standard.
In this blog post, I will take tuning data from the official Sonic Youth website and apply data science techniques in order to visualise and explore these alternate tunings, to look at their development with a view to understanding how they contributed towards their overall tonal characteristics.
Firstly, I would like to make clear that the alternate tunings were not the main focal point of the band. In an MTV 120 Minutes special from 1991 Lee Ranaldo stressed that it is just the way that they work, but it’s not what’s important about what they do. In the same interview Thurston Moore added “It’s still just songwriting”. I can understand why they would make a point of this because it would be easy for people to get hung up on the tunings and for them to be seen as a novelty act, just the band with the odd tunings when in actual fact they were accomplished songwriters with a strong experimental drive. That’s not to say the tunings should be dismissed. Anyone who has ever heard a familiar piece of music transposed to a different key will understand how much it can change the mood, which is why metal guitarists will detune their guitars, in order to give them a more sombre feeling.
For those unfamiliar with the guitar, the standard six string tuning, know as E Standard is tuned from string 6, the thickest string to string 1, the thinnest, using the notes E2, A3, D3, G3, B4, E4. The numbers here representing the octave numbers. This tuning can be visualised as such.
The blue bars represent the tuning of the open strings, the red the full range of notes available on a 22-fret guitar. As can be seen, the tuning of the strings are evenly spaced out in fourth intervals, with the exception of B4 which is tuned to a major-third. I will use E Standard tuning as a baseline to compare Sonic Youth’s tunings by calculating a relative numeric score. This will be done by taking the index number of the range between F1 and G#6 in the above which would be 12 for E2, and subtracting the difference from E2 which would be 12 in this case giving a score of zero. If a tuning was tuned down to C2 on string 6 the calculation would be 8 (index) - 12 (E2’s index) = -4. These differences can be calculated per string or summed to provide a single number meaning that E Standard has a total score of zero. I will then use this score to evaluate other tunings in order place them above or below in the frequency spectrum.
Now that we are familiar with the standard tuning, we can look at Sonic Youth’s tunings. Over the course of their career, they released fifteen studio albums, containing one-hundred and sixty-one songs, using a total of sixty-two different guitar tunings. Of these sixty-two there were some that would be reused on multiple songs as can be seen below.
The most used tuning was F#2, F#2, F#3, F#3, E3, B4, and what is instantly noticeable about this tuning are the doubled strings with a large interval of one octave between them, along with the shorter range compared to E Standard. To visualise it in the same way would look as so.
The sixth string is tuned upwards one tone from E2 to F#2 giving a brighter feel, and the second string E3 is lower than the third F#3, which is strange because usually as a guitar is strummed downwards each subsequent note is higher than the last. Using the aforementioned scoring system this tuning rates as -10. Even though the string 6 has been tuned up it has both a shorter and lower overall range than E Standard.
Looking at the second most used tuning G2, G2, D3, D3, D#3, D#3 we can see the pattern of doubled strings and a shorter range taken further which rates as -25 in the scoring system. Rather than the standard tuning, which was designed to be as versatile as possible, these tunings appear to be very specific, to allow for easy access to a particular range in the frequency spectrum. This is an interesting point which I’ll return to later.
Relating the tunings to the albums. If we look at the collective tunings for each album represented by colour we can see how the tunings have changed between each album. The x-axis represents each string, the y-axis the albums from first at the top to last at the bottom, with the colour representing the tuning of each string.
It’s pretty clear, even though they had favourite tunings, that they varied the tunings quite a lot between albums. We can see a drop to lower tunings in the middle period between Goo and NYC Ghosts & Flowers, while the last two albums seem to be tonally similar. A note about the NA that appears in the legend, this is due to a couple of tunings where strings four and three were not strung and were used by Thurston Moore on the songs Marilyn Moore and Eric’s Trip from the EVOL and Daydream Nation albums respectively.
If we use the scoring system to compare album tunings, but keep the string separation we can see where they lie in relation to E Standard, which makes differentiating a lot clearer. A score of zero, the most similar to E Standard is the darkest blue colour with higher tunings being purple/red and lower being brown/orange.
We can really see here the shortening of the range by the detuning of strings 2 and 1, especially in the mid-career period, except curiously on string 4 which was tuned up, the tuning up of that string also appearing on the earlier albums.
Looking at a box and whisker plot of the each string’s tunings, we can see the ranges involved. For those unfamiliar with this plot it shows the median and interquartile range, The thick horizontal line represents the median, the midpoint of the distribution which is called the second quartile. The box beneath the median shows the data down to the 25th percentile or the first quartile, this is where between 25% and 50% of the data are distributed. The box above the median shows the data up to the 75th percentile, the third quartile where between 50% and 75% of the data are distributed. The lines are the whiskers which show the upper and lower sets of each 25% of the data with the minimum and maximum values, while the dots represent outliers, isolated values.
It is interesting to see that the median of string 6 is at zero, with the larger third quartile showing a preference to tune this string upwards. Both strings 5 and 4 have no visible third quartile meaning that the upper 50% to 75% of that data lie on the median. Both these strings were largely tuned downwards or remained tuned to A3 in string 4’s case, and where detuned, by small amounts as shown by the smaller interquartile range. They both also have a lot of outliers which shows that Sonic Youth would sometimes break out of their tuning patterns, particularly with the these two strings. Strings 3 to 1 all have a median lower than zero with a larger first quartile showing how they would shorten the range of their guitars especially by tuning these string down by larger amounts than the others.
As Sonic Youth had two main guitarists, we cannot assume they always used the same tuning on the same song. Bassist Kim Gordon also played guitar on some album tracks before switching to guitar in the later part of their career, as did multi-instrumentalist Jim O’ Rourke who joined the band for two albums on which he played guitar on certain tracks. Breaking down the colour map further by guitarist we can see where in the frequency spectrum each person was placed.
Kim, having played bass, unsurprisingly seems to place lower in the spectrum while Jim places higher, with Thurston generally playing higher than Lee. We can check this quantitively by plotting the mean score for each guitarist’s tuning for each the albums.
This gives a more detailed picture. Kim does play lower on four albums, but higher afterwards. Lee plays lower than Thurston early on, but this changes and alternates with Jim not playing as high as first thought. From Dirty onwards we can see the mid-period where the tunings became lower, whereas before that they were not too dissimilar, while after the mid-period they became lower and more homogenous again. At no point in their career did they tune above E Standard suggesting that they alway like to shorten the range of the guitar, in effect customising their tunings specifically to the tonal range where they wanted the song to lie.
Going back to the most frequently used tunings, I wanted to see if they were used across all of the albums or were they only limited to certain periods? To look at this, I’ve taken the tunings used more than ten times and plotted their usage across all of the albums.
There seems to be a definite mid-career change after Washing Machine where some new tunings appear and the favourite of F#2, F#2, F#3, F#3, E3, B4 was largely dropped, only being used once more on Sonic Nurse. The second favourite of G2, G2, D3, G3, G3, A4 did get used on Murray Street and Sonic Nurse, but except that and minor usage of a couple of others, it seems they started preferring newer tunings after Washing Machine.
Bearing this in mind and expanding the scoring system visualisation into the individual songs across their whole career, we can see the ebb and flow of these patterns, as well as who contributed guitar towards each song.
Near the beginning of this post I mentioned that Sonic Youth’s tunings looked to be very specific and seemed to be designed to allow for easy access to a particular frequency range. This complements Lee Ranaldo’s comment that it was just the way that they work, and from the data this can be seen. It shows that they always reduced the range of their guitars to give them more of a narrow focus, to place them within the tonal spectrum they wished to use. There were some patterns emerging, particularly the mid-career lowering of the tunings with a switch to newer tunings that came afterwards, but what is clear from the amount of tunings used and the variability of them is that instead of using a versatile tuning such as E Standard and adapting their playing styles to get the results that they wanted, they worked the other way around and adapted the tunings to their playing styles. This fits in with the way that they would also customise their guitars. Instead of using stock configurations, they would replace the pickups, rewire the electronics, and jam screwdrivers between the strings at certain frets taking influence from John Cage’s prepared piano, and what this allowed them to do was to sculpt a unique sound that was entirely their own.
All of the visualisations were made using R and the tidyverse package. The data and the commented code can be found on my GitHub, and I’ve also made available a Jupyter Notebook.