Figure 15. Evocative stimuli (right column) are usually made with people, whereas nonevocative stimuli (left column) are more physics-related and sterile.
Do we find that evocativeness springs from the same human source within the auditory domain? Let’s start with speech. As we discussed in the previous chapter, speech sounds like solid-object physical events. “Solid-object physical events” amount to a sterile physics category of sound, akin in nerdiness to “three-dimensional world of opaque objects.” We are capable of mimicking lots of nonhuman sounds, and speech, then, amounts to yet another mimicry of this kind. Ironically, human speech does not sound human at all. It is consequently not evocative. (See the bottom left square of Figure 15 for speech’s place in the table.) Which brings us back to music, the other major kind of auditory stimulus people produce besides speech. Just as color is evocative but writing is not, music is evocative but speech sounds are not. This suggests that, just as color gets its emotionality from people, perhaps music gets its emotionality from people. Could it be that music, like Soylent Green, is made out of people? (Music has been placed at the bottom right of the table in Figure 15.)
If we believe that music sounds like people, then we greatly reduce the range of worldly sounds music may be mimicking. That amounts to progress: music is probably mostly not about birdsong, wind, water, math, and so on. But, unfortunately, humans make a wide variety of sounds, some in fundamentally different categories, such as speech, coughs, sneezes, laughter, heartbeats, chewing, walking, hammering, and so on. We’ll need a more specific theory than one that simply says music is made from people. Next, though, we ask why there isn’t any purely visual domain that is as exciting to us as music.
Going Solo
If the visual system and auditory system had competitive streaks, they might argue about which modality has the most compelling art. Each would be allowed to cite as examples only cases exclusively within its own modality: vision-only versus audition-only. This is a difficult contest to officiate. Should vision be allowed to cite all the features of visual design found in culture, such as clothes, cars, buildings, and everyday objects? If so, it would have a big leg up on audition, which is not nearly so involved in the design of our physical artifacts. Let’s agree not to include these, by virtue of an “official rule” that the art must be purchased by people for the purpose merely of enjoying the aesthetics, with no other functional benefit. That is, is it vision or audition that commands the greatest portion of the market for art and entertainment?
If you set it up in this way, audition trounces vision. Although the visual modality is found in huge markets like television, video games, and movies, these rely on audition as well. People put visual art on their walls, but that typically amounts to just a few purchases, whereas it is common to find people who own thousands of music albums. The market for the purely visual arts is miniscule compared to that for audition. This is counterintuitive, because if you ask most of us to name the most beautiful things we know of, we are likely to respond with a list of visuals. But when we vote with our pocketbooks, audition wins the solo artist contest. Why is that?
One possible explanation is simply that it is easier to carry on with the chores of life while music is in the background, whereas the visual arts inherently get in the way. Try driving or working or throwing a dinner party while admiring the Mona Lisa. But I suspect it is more than this. If it was merely because of the difficulty of enjoying visual arts while having a life, one might expect us to want to stare at beautiful visual art all day, if only we had nothing else pressing to do. Most of us, however, don’t exactly fancy the idea of watching visual images all day (without sound). Listening to music all day, however, sounds quite charming! And, in fact, many of us do spend our days listening to music.
The stark inequality of vision and audition in this competition for “best solo performer” in the arts is due to a fundamental ecological asymmetry. When we see things in the world, those things are typically making noise. Seeing without hearing therefore feels strange, unnatural, or as if it is missing something. But hearing without seeing is commonplace, because we hear all sorts of things we cannot see—when our eyes are closed, when the source is behind us, when the source is occluded, or when the environment is dark. Sights nearly always come with sounds, but sounds very commonly come without sights. And that’s why audition is happy to be a solo artist, but vision isn’t. Music is the single-modality artist extraordinaire.
While we now have some idea why there’s no solely visual art that rivals music, we still have barely begun our quest to understand why music is so compelling that we are willing to purchase thousands of albums.
At the Heart of a Theory of Music
If music sounds human in fundamental respects, as our discussion in the section before last suggested, then it seems to have made heroic efforts to obfuscate this fact. I readily admit that music doesn’t sound human to me—not consciously, at least. But recall the section titled “Below the Radar” from Chapter 1, where I said that we don’t necessarily expect cultural artifacts to mimic nature “all the way up.” It may be the case that much of our lower-level auditory apparatus thinks that music sounds like humans, but that because of certain high-level dissimilarities, we—our conscious selves—don’t notice it. How, then, can I hope to convince anyone? I have to convince you, after all, not your lower-level auditory areas!
What we need are some qualifying hurdles that a theory of music should have to leap over to gain a hearing . . . hurdles that, once cleared, will serve to persuade some of Earth’s teeming music buffs that music does indeed sound like people moving. Toward this end, here are four such hurdles—questions that any aspiring theory of music might hope to answer.
Brain: Why do we have a brain for music?
Emotion: Why is music emotionally evocative?
Dance: Why do we dance?
Structure: Why is music organized the way it is?
If a theory can answer all four questions, then I believe we should start paying attention.
To help clarify what I mean by these questions, let’s run through them in the context of a particular lay theory of music: the “heartbeat” theory. Although there is probably more than just one heartbeat theory held by laypeople, the main theme appears to be that a heart has a beat, as music does. Of course, we don’t typically hear our own heartbeat, much less others’, so when the theory is fleshed out, it is often suggested that the fundamental beat was laid down when we were in utero. One of the constants of the good fetal life was Momma’s heartbeat, and music takes us back to those oceanic, one-with-the-universe feelings we long ago lost. I’m not suggesting that this is a good theory, by any means, but it will aid me in illustrating the four hurdles. I would be hesitant, by the way, to call this “lub-dub” theory of music crazy—our understanding of the origins of music is so woeful that any nonspooky theory is worth a look. Let’s see how lub-dubs fare with our four hurdles for a theory of music.
The first hurdle was this: “Why do we have a brain for music?” That is, why are our brains capable of processing music? For example, fax machines are designed to process the auditory modulations occurring in fax machine communication, but to our ears fax machines sound like a fairly continuous screech-brrr—we don’t have brains capable of processing fax machine sounds. Music may well sound homogeneously screechy-brrrey to nonhuman ears, but it sounds richly dynamic and structured to our ears. How might the lub-dub theorist explain why we have a brain for music? Best I can figure, the lub-dubber could say that our in-utero days of warmth and comfort get strongly associated to Momma’s heartbeat, and the musical beat taps into those associations, bringing back warm fetus feelings. One difficulty for this hypothesis is that learned associations often don’t last forever, so why would those Momma’s-heartbeat associations be so strong among adults? There are lots of beatlike stimuli outside of the womb: some are nice, some are not nice. Why wouldn’t those out-of-the-womb sounds become the dominant ass
ociations, with Momma’s heartbeat washed away? And if Momma’s lub-dubs are, for some reason, not washed away, then why aren’t there other in utero experiences that forever stay with us? Why don’t we, say, like to wear artificial umbilical cords, thereby evoking recollections of the womb? And why, at any rate, do we think we were so happy in the womb? Maybe those days, supposing they leave any trace at all, are associated with nothing whatsoever. (Or perhaps with horror.) The lub-dub theory of music does not have a plausible story for why we have a brain ready and eager to soak up a beat.
The lub-dub theory of music origins also comes up short in the second major demand on a theory of music: that it explain why music is evocative, or emotional. This was the subject of the previous section. Heartbeats are made by people, but heartbeat sounds amount to a one-dimensional parameter—faster or slower rate—and are not sufficiently rich to capture much of the range of human emotion. Accordingly, heartbeats won’t help much in explaining the range of emotions music can elicit in listeners. Psychophysiologists who look for physiological correlates of emotion take a variety of measurements (e.g., heart rate, blood pressure, skin conductance), not just one. Heart sounds aren’t rich enough to tug at all music’s heartstrings.
Heartbeats also fail the “dance” hurdle. The “dance” requirement is that we explain why it is that music should elicit dance. This fundamental fact about music is a strange thing for sounds to do. In fact, it is a strange thing for any stimulus to do, in any modality. For lub-dubs, the difficulty for the dance hurdle is that even if lub-dubs were fondly recalled by us, and even if they managed to elicit a wide range of emotions, we would have no idea why they should provoke post-uterine people to move, given that even fetuses don’t move to Momma’s heartbeat.
The final requirement of a theory of music is that it must explain the structure of music, a tall order. Lub-dubs do have a beat, of course, but heartbeats are far too simple to begin to explain the many other structural regularities found in music. For starters, where is the melody?
Sorry, Mom (again). Thanks for the good times in your uterus, but I’m afraid your heartbeats are not the source of my fascination with music.
Although the lub-dub theory fails the four requirements for a theory of music, the music-sounds-like-human-movement theory of music, as we will see, has answers to all four. We have a brain for music because possessing auditory mechanisms for recognizing what people are doing around us is clearly advantageous. Music is evocative because it sounds like human behaviors, many of which are expressive in their nature—something we will discuss further in a few pages. Music gets us dancing because, as we will also discuss, we social apes are prone to mimic the movements of others. And, finally, the movement theory is sufficiently powerful that it can explain a lot of the structure of music—that will require the upcoming chapter and the Encore (at the end of the book) to describe.
Underlying Overtones
The heartbeat theory suffered cardiac arrest, but it was never intended as a serious contender. It was just a prop for illustrating the four hurdles. Speech, on the other hand, is a much more plausible starting point as a foundation for music. But haven’t we already discussed speech? Wasn’t that what the previous chapter was about? We concluded then that speech sounds like solid-object physical events: the structural regularities found among solid-object events are reflected in the phonological patterns of human speech. Speech is all about the phonemes, and how closely they mimic nature’s pattern of hits, slides, and ring sounds. Music, on the other hand, cares not a whit for phonemes. Although music can often have words to be sung, music usually gets its identity not from the words, but from the rhythm and tune. Two songs with different words, but with the same rhythm and pitch sequence, are deemed by us to be the same tune, just with different words. That’s why we use the phrase “put words to the music”—because the words (and the phonemes) are not properly part of the meat of the music. The most central auditory feature of speech—its phonological characteristics—is mostly irrelevant to music, making speech an unlikely place to look for the origins of music.
Music is not only missing the phonological core of speech, but it is also missing another fundamental aspect of speech, its most evocative aspect: the meaning, or semantics. If music has its source in speech, and is evocative because of the evocative nature of speech, then why wouldn’t music require words with meaning, whether metaphorical or direct? Yet, as mentioned above, neither phonology nor words is an essential ingredient of music. (Although phonology and words are key ingredients in poetry.)
If music comes from speech, then it doesn’t come from the phonological patterns of speech, or from the semantics of speech. Although these core functions of speech are dead ends for a theory of music, there is another aspect of speech I have purposely glossed over. People overlay the sterile solid-object event sounds of speech with emotional overtones. We add intonation, a pitchlike property. We vary the emphasis of the words in a sentence, reminiscent of the way rhythm bestows emphasis in music (for instance, the first beat in a measure usually has enhanced emphasis). We vary the timing of the word utterances, akin to the temporal patterns of rhythm in music. And we sometimes modulate the overall loudness of our voices, like a musical crescendo or diminuendo. These prosody-related emotional overtones turn Stephen Hawking computer-voice speech into regular human speech. And these emotional overtones can be understood even in foreign speech, where our ears can often recognize the glib, the mournful, the proud, and the angry. We’re just not sure what they are glib, mournful, proud, or angry about.
So it is not quite true that speech sounds are sterile. Rather, it is the phonological solid-object event sounds that are sterile. The overtones of speech, on the other hand, are dripping with human emotion. Might these overtones underlie music? In an effort to answer, let’s discuss the four questions at the heart of any theory of music, the ones I referred to earlier as “brain,” “emotion,” “dance,” and “structure.”
Do we have a brain for the overtones of speech? An overtone theory of music would like to say that music “works” on our brains because it taps into speech overtone recognition mechanisms. Are we likely to have neural mechanisms for recognizing overtones of speech? Although I am suggesting in this book that we did not evolve to possess speech recognition mechanisms, we primates have been making nonspeech vocalizations (cries, laughs, shrieks, growls, moans, sighs, and so on) for tens of millions of years, and surely we have evolved neural mechanisms to recognize them. Perhaps the overtones of speech come from our ancient nonspeech vocalizations, and they get laid on top of the solid-object physical event sounds of speech like a whipped cream of evocativeness, a whipped cream our auditory system knows how to taste. An overtone-based theory of music, then, does have a plausible story to tell about why our brain would be highly efficient at recognizing overtones.
Can overtones potentially explain the evocativeness of music, the second hurdle we had discussed for any theory of music? Of course! Overtones are emotional, used in vocalization to be evocative. If music mimics emotional overtones, then it is easy to grasp how music can be evocative.
Can an overtone theory of music explain dance, the third hurdle I mentioned earlier? One can see how the emotional nonspeech vocalization of other people around us might provoke us into action of some kind—that’s probably why people are vocalizing in the first place. That’s a start. But we would like to know why hearing overtones would not just tend to provoke us to do stuff, but more specifically, make us move in a time-locked fashion to the emotional vocalizations. I have not been able to fathom any overtone-related story that could explain this, and the absence of any potential connection to dance is a hurdle that an overtone theory stumbles over.
Finally, can overtones explain the structure of music? Do the overtones of speech possess the patterns of pitch, loudness, and rhythm found in music? There is, at least, enough structure floating around in the prosody of speech that one can imagine it might be rich enough to help
explain the structure found in music. But despite the nice confluence between ingredients in the overtones of speech and certain similar ingredients in music, overtones appear to be a very different beast from music. First and foremost, what’s missing in the overtones of speech is a beat, and a rhythm time-locked to a beat. That’s the one thing the lub-dub theory of music captured, but it is one of the most glaring shortcomings of overtone-based approaches, and it ultimately takes overtones of speech out of the running as a basis for a theory of music.
Before leaving speech for more fertile grounds—in fact, the next section is about sex—consider the two hurdles where overtones appeared promising: “brain” and “emotion.” I suggested earlier in this section that overtones could rely on ancient human nonlinguistic vocalizations, but there is another potential foundation for overtones’ evocative nature: the sounds of people moving. Rather than music coming from the overtones of speech, perhaps both music and overtones have their foundation in the more fundamentally meaningful sound patterns of humans’ expressive movements. (And perhaps this is the source of the intersections between music and speech in the brain discerned by Aniruddh D. Patel of the Neurosciences Institute in La Jolla, and other researchers.)
Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man Page 10