Our first knee jerk reaction is that the word “music” and the word “distortion” should not be uttered in the same sentence. Music is good (or at least should be of high fidelity) and distortion is, well,… distortion.
But if it weren’t for distortion, music would be thin and boring, and speech would be completely unintelligible. Without distortion, music would not be music. Norman Greenbaum would never have been able to write Spirit in the Sky and introduce an entire generation to the beauty of the sound of a broken loudspeaker cone (at least that is the rumor). And Jacky Brenston and the Delta Cats would never have been able to make their great recording of Rocket ’88 in 1951 using a broken amplifier that defined rock and roll.
The romantic story is that Norman Greenbaum dropped his loudspeaker on the way to a gig and Spirit in the Sky sounds the way it does because a torn loudspeaker cone imparted a great distorted sound. In fact, in chatting with Norman several days ago, he said although he is in favor of anything romantic, alas this story is untrue. From the start he used a fuzz generator with a lot of reverberation, and he played without a pick.
In the case of Rocket ’88, the legendary Sam Phillips of Sun Records in Memphis had a four-vacuum tube (“valve” for our readers from the United Kingdom) amplifier in his studio that only had three functioning tubes. The three remaining amplification stages were maximized in order to offset the missing stage and this worked quite well…except with the more intense sounds. With intense sounds it distorted significantly, giving a bass sound that was not actually in the input. The Delta Cats did not have a bass player but audiophiles to this day swear that they hear a bass. The “missing bass” was created by low-frequency intermodulation distortion.
And let’s talk about speech. The output from the vocal tract creates significant harmonic distortion. My fundamental frequency (pitch) is about 125 Hz, but when I utter a vowel sound, not only is there energy at 125 Hz but also integer multiples of 125 Hz- 250 Hz, 375 Hz, 500 Hz, 625 Hz, 750 Hz, and so on. This pattern of integer multiples of the fundamental frequency is consistent with a half wavelength resonator.
A half wavelength resonator can be constructed from any structure that has the same acoustical and mechanical properties at both ends. Human vocal chords are held tightly at both ends and are free to vibrate freely in the center. A guitar is also a one half wavelength resonator- the two ends of the string are held tightly at both ends but are free to vibrate in the center. Using this same reasoning, violins, violas, and all other stringed instruments generate integer numbered harmonics of the fundamental. These harmonics are distortion.
Taking this a step further, one can demonstrate that the basilar membrane in the Organ of Corti in our cochlea is a distorting mechanism creating harmonic distortion, both of the summation type (higher frequencies) and of the intermodulation type (combination tones such as 2f1-f2). As many people age, the chemistry of the cochlea changes resulting in a more linear basilar membrane response with fewer distortion products. This loss of informational cues (the missing distortion products) may account, at least in part, for a senior citizen’s poorer functioning ability in noise if matched with a young person with the same hearing status.
In contrast to one half wavelength resonator musical instruments (strings, vocal, and some others such as the flute, saxophone, oboe, and bassoon) there are some one quarter wavelength musical instruments such as the brass instruments and the lower register of the clarinet (the upper register is a one half wavelength resonator… go figure). The one quarter wavelength resonator musical instruments have odd numbered multiples of the fundamental. If the fundamental was 125 Hz (such as that of my voice and which is about an octave below middle C on a piano keyboard), then the first harmonic would be 3 x 125 Hz, then 5 x 125 Hz, and so on. We would have a spectrum composed on 125 Hz, 375 Hz, 625 Hz, and so on.
Then we have unusual musical instruments such as the percussion that function, at least in part, as non-mathematically simply resonators- there may be distortion products that are not simple integer multiples of the fundamental. These distortion products are called overtones and, unlike harmonics, may not bear any simple relationship to the fundamental.
The bottom line is that distortion is the creation of well defined harmonics (and overtones) and it is these additional items that were not in the original music and speech signal that contributes to fidelity and intelligibility. Without these higher frequency distortion products (and potentially lower frequency intermodulation distortion products) music would not be music. And speech would not be speech.
Distortion is good, at least up to a point. Too much distortion may be as bad as too little distortion. Like many things in life, the correct balance may be just right. But sometimes, like Spirit in the Sky, and Rocket ’88, more may be better.