Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man

by Mark Changizi

  Rhythm and beat have, then, some similarities to the structure of our banging ganglies. We will discuss more similarities in the upcoming sections and in the Encore chapter. But there is one important similarity that might appear to be missing: musical notes usually come with a pitch, and yet our footsteps and gangly hits are not particularly pitchy. How can the dull thuds of our bodies possibly be pitchy enough to explain the central role of pitch in music?

  If you have already read the earlier chapter on speech, then you may have begun to have an appreciation for the rings occurring when any solid-object physical event occurs. As we discussed, we are typically not consciously aware of the rings, but our auditory system hears them and utilizes them to determine the identity of the objects involved in events (e.g., to tell the difference between a pencil and a paper clip hitting a desk). Although the pitch of a typical solid object may not be particularly salient, it can become much more salient when contrasted with the distinct pitches of other objects’ rings. For example, a single drum in a set of drums doesn’t sound pitchy, but when played in combination with larger and smaller drumheads, each drum’s pitch becomes easy to hear. The same is true for percussionists who use everyday objects for their drums—in such performances one is always surprised to hear the wide range of pitches occurring among all the usually pitchless-seeming everyday objects. Our footsteps and banging ganglies do have pitches, consistent with the hypothesis that they are the fundamental source of musical notes. (As we will see, these gangly pitches are analogous to chords, not to melody—which, I will argue later, is driven by the Doppler effect.)

  If I am right that musical notes have their origin in the sounds that humans make when moving, then notes should come in human-gait-like patterns. In the next section, we’ll take up a simple question in this regard: does the number of notes found between the beats of music match the number of gangly bangs between footsteps?

  The Length of Your Gangly

  Every 17 years, cicadas emerge in droves out of the ground in Virginia, where I grew up. They climb the nearest tree, molt, and emerge looking a bit like a winged tank, big enough to fill your palm. Since they’re barely able to fly, we used to set them on our shoulders on the way to school, and they’d often not bother to fly away before we got there. And if they did fly, it wasn’t really flying at all. More of an extended hop, with an exoskeleton-shaking, tumble-prone landing. With only a few days to live, and with billions of others of their kind having emerged at the same time, all of them screeching mind-numbingly away, they didn’t need to go far to find a mate, and graceful flight did not seem to be something the females rewarded.

  Cicadas have, then, a distinctively cicada-like sound when they move: a leap, a clunky clatter of wings, and a heavy landing (often with further hits and skids afterward). The closest thing to a footstep in this kind of movement is the landing thud, and thus the cicada manages to fit dozens of banging ganglies—its wings flapping—in between its landings. If cicadas were someday to develop culture and invent music that tapped into their auditory movement-recognition mechanisms, then their music might have dozens of notes between each beat. With Boooom as their beat and da as their wing-flap inter-beat note, their music might be something like “Boooom-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-Boooom-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da-da,” and so on. Perhaps their ear-shattering, incessant mating call is this sound!

  Whereas cicadas liberally dole out notes in between the beats, Frankenstein’s monster in the movies is a miser with his banging ganglies, walking so stiffly that his only gait sounds are his footsteps. Zombies, too, tend to be low on the scale of banging-gangly complexity (although high on their intake of basal ganglia).

  When we walk, our ganglies are more complex than those of Frankenstein and his zombie dance buddies, but ours are doled out much more sparingly than the cicadas’. During a step, your leg swings forward just once, and so it can typically only get one really good bang on something. More complex behaviors can lead to more bangs per step, but most commonly, our movements have just one between-the-footsteps bang—or none. Our movements tend to sound more like the following, where “Boooom” is the regularly repeating footstep sound and “da” is the between-the-steps sound: “Boooom-Boooom-Boooom-da-Boooom-Boooom-da-Boooom-da-Boooom-da-da-Boooom-da-Boooom-da-Boooom.” (Remember to do the “Boooom” on the beat, and cram the “da”s in between the beats.)

  Given our human tendency to make roughly zero to one gangly bang between our steps, our human music should tend to pack notes similarly lightly between the beats. Music is thus predicted to tend to have around zero to one between-the-beats note. To test for this, we can look at the distribution of time gaps between musical notes. If music most commonly has about zero to one note between the beats—along with notes usually on the beat—then the most common note-to-note time gap should be in the range of a half beat to a beat.

  To test this, as an RPI graduate student, Sean Barnett analyzed an electronic database of Barlow and Morgenstern’s 10,000 classical themes, the ones we mentioned at the start of this chapter. For every adjacent pair of notes in the database, Sean recorded the duration between their onsets (i.e., the time from the start of the first note to the start of the second note). Figure 19 shows the distribution of note-to-note time gaps in this database—which time intervals occur most commonly, and which are more rare. The peak occurs at ½ on the x-axis, meaning that the most common time gap is a half beat in length (an eighth note). In other words, there is one note between the beats on average, which is broadly consistent with expectation.

  Figure 19. The distribution of durations between notes (measured in beats), for the roughly 10,000 classical themes. One can see that the most common time gap between notes is a half beat long, meaning on average about one between-the-beat note. This is similar to human gait, typically having around zero to one between-the-step “gangly” body hit.

  We see, then, that music tends to have the number of notes per beat one would expect if notes are the sounds of the ganglies of a human—not a cicada, not a Frankenzombie—mover. Musical notes are gangly hits. And the beat is that special gangly hit called the footstep. In the next section we will discuss some of what makes the beat special, and see if footsteps are similarly special (relative to other kinds of gangly hits).

  Backbone

  My family and I just moved into a new house. Knowing that my wife was unhappy with the carpet in the family room, and knowing how much she fancies tiled floor, I took the day off and prepared a surprise for her. I cut tile-size squares from the carpet, so that what remained was a checkerboard pattern, with hardwood floors as the black squares and carpet as the white squares.

  I couldn’t sleep very well that night on the couch, and so I headed into the kitchen for a bite. As I pondered how my plan had gone so horribly wrong, I began to notice the sounds of my gait. Walking on my newly checkered floor, my heels occasionally banged loudly on hard wood, and other times landed silently on soft carpet. Although some of my between-step intervals were silent, between many of my steps was a strong bump or shuffle sound when my foot banged into the edge of the two-inch-raised carpet. The overall pattern of my sounds made it clear when my footsteps must be occurring, even when they weren’t audible.

  Luckily for my wife—and even more so for me—I never actually checkered my living room carpet. But our world is itself checkered: it is filled with terrain of varying hardness, so that footstep loudness can vary considerably as a mover moves. In addition to soft terrain, another potential source of a silent step is the modulation of a mover’s step, perhaps purposely stepping lightly in order to not sprain an ankle on a crooked spot of ground, or perhaps adapting to the demands of a particular behavioral movement. Given the importance of human footstep sounds, we should expect that our auditory systems were selected to possess mechanisms capable of recognizing human gait sounds even when some
footsteps are missing, and to “fill in” where the missing footsteps are, so that the footsteps are perceptually “felt” even if they are not heard.

  If our auditory system can handle missed footsteps, then we should expect music—if it is “about” human movement—to tap into this ability with some frequency. Music should be able to “tell stories” of human movement in which some footsteps are inaudible, and be confident that the brain can handle it. Does music ever skip a beat? That is, does music ever not put a note on a beat?

  Of course. The simplest cases occur when a sequence of notes on the beat suddenly fails to continue at the next beat. This happens, for example, in “Row, Row, Row Your Boat,” when each “row” is on the beat, and then the beat just after “stream” does not get a note. But music is happy to skip beats in more complex ways. For example, in a rhythm like that shown in Figure 20, the first beat gets a note, but all the subsequent beats do not. In spite of the fact that only the first beat gets a note, you feel the beat occurring on all the subsequent skipped beats. Or the subsequent notes may be perceived to be off-beat notes, not notes on the beat. Music skips beats and humans miss footsteps—and in each case our auditory system is able to perceptually insert the missing beat or footstep where it belongs. That’s what we expect from music if beats are footsteps.

  Figure 20. The first note is on the beat, but because it is an eighth note (lasting only half a beat), all the subsequent quarter notes (which are a beat in length) are struck on the off beat. You feel the beat occurring between each subsequent note, despite there being no note on the beat.

  The beat is the solid backbone of music, so strong it makes itself felt even when not heard. And the beat is special in other ways. To illustrate this, let’s suppose you hear something strange approaching in the park. What you find unusual about the sound of the thing approaching is that each step is quickly followed by some other sound, with a long gap before the next step. Step-bang . . . . . . Step-bang . . . . . . “What on Earth is that?” you wonder. Maybe someone limping? Someone walking with a stick? Is it human at all?! The strange mover is about to emerge on the path from behind the bushes, and you look up to see. To your surprise, it is simply a lady out for a stroll. How could you not have recognized that?

  You then notice that she has a lilting gait in which her forward-swinging foot strikes the ground before rising briefly once again for its proper footstep landing. She makes a hit sound immediately before her footstep, not immediately after as you had incorrectly interpreted. Step . . . . . . bang-Step . . . . . . bang-Step . . . . . . Her gait does indeed, then, have a pair of hit sounds occurring close together in time, but your brain had mistakenly judged the first of the pair of sounds to be the footstep, when in reality the second in the pair was the footstep. The first was a mere shuffle-like floor-strike during a leg stride. Once your brain got its interpretation off-kilter, the perceptual result was utterly different: lilting lady became mysterious monster.

  The moral of this lilting-lady story is that to make sense of the gait sounds from a human mover, it is not enough to know the temporal pattern of gait-related hit sounds. The lilting lady and mysterious monster have the same temporal pattern, and yet they sound very different. What differs is which hits within the pattern are deemed to be the footstep sounds. Footsteps are the backbone of the gait pattern; they are the pillars holding up and giving structure to the other banging gangly sounds. If you keep the temporal pattern of body hits but shift the backbone, it means something very different about the mover’s gait (and possibly about the mover’s identity). And this meaning is reflected in our perception.

  If musical rhythm is like gait, then the feel of a song’s rhythm should depend not merely on the temporal pattern of notes, but also on where the beat is within the pattern. This is, in fact, a well-known feature of music. For example, consider the pattern of notes in Figure 21.

  Figure 21. An endlessly repeating rhythm of long, short, long, short, etc., but with neither “long” nor “short” indicated as being on the beat. One might have thought that such a pattern should have a unique perceptual feel. But as we will see in the following figure, the pattern’s feel depends on where the beat-backbone is placed onto it. Human gait is also like this.

  One might think that such a never-ending sequence of long-short note pairs should have a single perceptual feel to it. But that same pattern sounds very different in the two cases shown in Figure 22, which differ only in whether the short or the long note marks the beat. The first of these sounds jarring and inelegant compared to the second. The first of these is, in fact, like the mysterious monster we imagined approaching a moment ago, and the second is like the lilting lady the mover turned out to be.

  Figure 22. (a) A short-long rhythm, which sounds very different from (and less natural than) the long-short rhythm in (b).

  Music, like human gait-related sounds, cannot have its beat shifted willy-nilly. The identity of a gait depends on which hits are the footsteps, and, accordingly, the identity of a song depends on which notes are on the beat. And when a beat is not heard, the brain infers its presence, something the brain also does when a mover’s footstep is inaudible.

  There are, then, a variety of suspicious similarities between human gait and the properties of musical rhythm. In the upcoming section, we begin to move beyond rhythm toward melody and pitch. We’ll get there by way of discussing how chords may fit within this movement framework, and how choreography depends on more than just the rhythm.

  Although we’re moving on from rhythm now, there are further lines of evidence that I have included in the Encore, which I will only provide teasers for here:

  Encore 1: “The Long and Short of Hit” Earlier in this section I mentioned that the short-long rhythm of the mysterious monster sounds less natural than the long-short rhythm of the lilting lady. In this part of the Encore, I will explain why this might be the case.

  Encore 2: “Measure of What?” I will discuss why changing the measure, or time signature, in music modulates our perception of music.

  Encore 3: “Fancy Footwork” When people change direction while on the move, their gait often can become more complex. I show that the same thing occurs in music: when pitch changes (indicative, as we will see, of a turning mover), rhythmic complexity rises.

  Encore 4: “Distant Beat” The nearer movers are, the more of their gait sounds are audible. I will discuss how this is also found in music: louder portions of music tend to have more notes per beat.

  Gangly Chords

  Earlier in this chapter, we discussed how footsteps and gangly bangs ring, and how these rings tend to have pitches. I hinted then that it is the Doppler shifting of these pitches that is the source of melody, something we will get to soon in this chapter. But we have yet to talk about the other principal role of pitch in music—harmony and chords.

  When pitches combine in close temporal proximity, the result is a distinct kind of musical sound called the chord. For example, C, E, and G pitches combine to make the C major chord. Where do chords fit within the music-is-movement theory? To begin to see what aspect of human movement chords might echo, consider what happens when a pianist wants to get a rhythm going. He or she could just start tapping the rhythm on the wood of the piano top, but what the pianist actually does is play the rhythm via the piano keys. The rhythm is implemented with pitches. And furthermore, the pianist doesn’t just bang out the rhythm with any old pitches. Instead, the pianist picks a chord in which to establish the rhythm and beat. What the pianist is doing is analogous to what a guitarist does with a strum. Strums, whether on a guitar or a piano, are both rhythm and chord.

  My suspicion is that rhythm and chords are two distinct kinds of information that come from the gangly banging sounds of human movers. I have suggested in this chapter that rhythm comes from the temporal pattern of human banging ganglies. And now I am suggesting that chords come from the combinations (or perhaps the constituents) of pitches that occur among the banging gangly rings.
Gait sounds have temporal patterns and pitch patterns, and these underlie rhythm and chords, respectively. And these two auditory facets of gait are informative in different ways, but both broadly within the realm of “attitude” or “mood” or “intention,” as opposed to being informative about the direction or distance of the mover—topics that will come up later in regard to melody and loudness, respectively.

  If rhythm and chords are each aspects of the sounds of our ganglies, then we should expect chords to cycle through their pitches on a time scale similar to that of the rhythm, and time-locked to the rhythm; the rhythm and chord should have the same time signature. For example, in an Alberti chord/rhythm pattern, one’s left hand on the piano might play the notes [CGEG][CGEG][CGEG], where each set of square brackets shows a two-beat interval, and bold type and underlines indicate the emphases in the rhythm. One can see that the same two-beat pitch pattern and rhythm repeats over and over again. The pitch sequence and the rhythm have the same 2/4 time signature. It is much rarer to find chords expressed in a way that mismatches the rhythm, such as the following case, where the chord is expressed as a repeated pattern of three pitches—C-G-E—and thus the two-beat rhythm cycles look like [CGEC][GECG][ECGE]. In this case, notice that the first two-beat interval—the rhythm’s cycle—has the pitch sequence CGEC, but that the second one has, instead, GECG. The pitch cycle for the chord is not matched to the rhythm’s cycle. In real music, if the rhythm is in 2/4 time, then the chord will typically not express itself in ¾ time. Rhythm and chords tend to be locked together in a way that suggests they are coming from the same worldly source, and therefore the arguments in this chapter lead one to speculate that both rhythm and chords come from, or are about, our gangly banging sounds.