The World in Six Songs: How the Musical Brain Created Human Nature

by Levitin, Daniel J.

  Interestingly, this cognitive leap that brought about art didn’t follow a sudden change in brain size. All three features of the musical brain—perspective taking, representation, and rearrangement—arise from modes of thought that are surprisingly similar in other animals, and differences in brain anatomy that are quantifiable but not especially dramatic. This is because human mental abilities are built from existing structures found in other animals that show many similar features. In other words, there exists a continuum of mental skills in the animal kingdom, and in virtually every case of an ability, it is not that we humans have a wholly different ability (a difference of kind), rather, it is that we have more ability (a difference of degree). Darwin himself noted this.

  How and why perspective taking, representation, and rearrangement came about in human brains we still don’t know, but the proximal cause of the difference was probably unexceptional. In other words, small adaptations in the prefrontal cortex, perhaps too small to detect from the fossil record, allowed neural circuitry to cross a threshold of complexity that bestowed on our lucky ancestors the brain power to make these significant cognitive leaps. But the unexceptionalness of the biological change (the cognitive result is quite exceptional) can be understood by appreciating how much of what we can do can also be done by other species.

  Take language communication as an example. For years it has been argued that many species have forms of communication, but only humans have language. Scientists state very specific demands for what constitutes language and what doesn’t, and on the surface, the way that we humans communicate with one another does exhibit several key features that are different from what animals do. Among those features, there are no reported cases of animals naming things spontaneously. Some chimps and apes have succeeded in learning sign language, and dogs can learn to label objects differentially (my dog Shadow distinguishes several different toys by name including his “fuzzy man” from his “Cat in the Hat”), but this only demonstrates their ability to link a visual or acoustic stimulus with an object. There is no evidence that these animals understand that the name refers to the thing (that it has intensionality), only that they have associated the sign or sound and the object in a kind of Pavlovian, unconscious connection. Moreover, once they’ve learned to name things, animals don’t teach other animals to name things.

  Another feature of human language is that it can be used reflexively. We talk about our language, discuss whether we’re using a word appropriately, we make a distinction between whether we’re relaying the gist of a conversation or quoting a person verbatim. We make “quote” symbols in the air with our fingers to indicate when we’re using a phrase literally or ironically, and we use the music of language—prosody—to help distinguish such literalness from irony.

  A third important quality of human language is that it is infinitely expandable—every day humans produce or hear utterances that have never before been spoken, and yet we understand them. Human language is expandable in two ways. First, there is no such thing as the longest sentence in the English language (or any other language we know of), because whatever sentence you nominate for the longest, I can always make it longer by adding a clause to the front of it, such as “David Bowie thinks that . . .” So “My dog has fleas” becomes “David Bowie thinks that my dog has fleas.” Or the longish sentence, “The album Brothers in Arms by Mark Knopfler and Dire Straits, in addition to containing great songwriting and performances, may well be one of the most perfectly engineered and mixed albums in the history of recorded sound” becomes “David Bowie thinks that the album Brothers in Arms . . .” (and so on). In this respect, human language is like the natural numbers—there is no number larger than all the others, because I can always add the number 1 (or any other positive number) to the number you nominate as the largest. (In fact, mathematicians have found utility in talking about the difference between ∞ and ∞ + 1.)

  The other way that human languages are expandable is that words can be combined in different ways. No child learns a list of all the possible sentences he will need; rather, he learns words, and then rules about how new words are formed and about how they can be combined. This comes under the musical brain’s capacity for abstraction—we know that the words are merely elements of an utterance that stand for specific things, and that they can be recombined or substituted to change the meanings of utterances. “The cat chased the dog” means something different than “The dog chased the cat” by virtue of the ordering of the words. (Grammar specifies, among other things, rules about who is doing what to whom based on where in the sentence the elements appear.) We can say “I am going to the park ” or “I am going to the zoo” using similar form and structure, but different words.

  Animals use fixed, irreducible phrases. The black-capped chickadee of North America has thirteen distinct vocalizations, but there is no evidence that they can be combined, or that part of one can be substituted or inserted in another. The “chick-a-dee-dee-dee” call is the whole message. “Chick-a-dee-dee-doo” or “Chuck-ee-dee-dee-dee” are not possible messages in chickadee language.

  Steven Mithen argues in his book The Singing Neanderthals that Neanderthals may have communicated using a kind of protomusic with pitch, rhythm, timbral, and loudness variations. Moreover, Mithen believes they employed a repertoire of fixed calls, constituting something slightly more sophisticated than those in use by crows and vervet monkeys, for communicating such holistic thoughts as “Look out! There’s a snake!” and “Come get some food.” Unlike human languages, in which words can be substituted in sentences to alter their meanings (“Come get some water”), the holistic Neanderthal utterances would have been more like monkey or bird calls—fixed and not extendable or changeable. They were neither language nor music as we think of them today. Darwin believed that music as we know it is a kind of fossil, a remnant of an earlier communication system or “musical protolanguage.”

  Noam Chomsky, the Hunter S. Thompson of linguists, proposed that language may be decomposable into two components—conveyance and computation. The distinction maps to two different, historically sequential forms of language. The first kind to appear was an unstructured protolanguage that conveyed concepts, meaning, and emotions, like Mithen’s Neanderthal language, and like contemporary chimpanzee, monkey, and gibbon communication (and presumably like that of Australopithecines). The conveyance form of language served to communicate the right-now : the emotional state of the communicator, the existence of a predator or of food, and so on. The second kind arose later, when the human brain evolved computational modules capable of rearranging elements to a preconceived plan, and of imposing a complex hierarchy on objects, including linguistic objects; in the case of utterances, this constitutes a structure and order we call syntax. (I prefer the term rearrangement to the Chomsky’s computation, and so I’ll use that here.) This ability to understand, form, and analyze hierarchies is a key development that led both to music and to language. Chomsky doesn’t say so explicitly, but it seems that along with this hierarchical processing came the ability to communicate about things beyond the right-now—to talk about things that happened in the past or might happen in the future.

  I believe that Chomsky’s conveyance and computational (rearrangement) system has parallels in religion, love, and music. As we saw in Chapter 6, nearly all religious acts involve repetitive, stereotyped motor action sequences—rituals. The distinction to be made here is that all primates engage in ritual—a way of displaying (or, in Chomsky’s terminology, conveying) an emotional state. The ritual is the conveyance system. Religion adds to ritual a computational component—the ability to recombine, conceptualize, and recontextualize the ritual—to give it meaning and order. Human love is more than the ability to show attachment; it entails thinking about a hierarchy of importance, to plan to be in love in the future, and to communicate this plan to others.

  Darwin, in The Descent of Man, spoke about animal “love,” and it is clear from this that he is referri
ng to conveyance and to attachment rather than to the (human) computational component:Animals of many kinds are social; we find even distinct species living together; for example, some American monkeys; and united flocks of rooks, jackdaws, and starlings. Man shews the same feeling in his strong love for the dog, which the dog returns with interest. Every one must have noticed how miserable horses, dogs, sheep, &c., are when separated from their companions, and what strong mutual affection the two former kinds, at least, shew on their reunion. It is curious to speculate on the feelings of a dog, who will rest peacefully for hours in a room with his master or any of the family, without the least notice being taken of him; but if left for a short time by himself, barks or howls dismally.

  Later, Darwin discusses the evolutionary origin of attachment, which led to what we think of as human love:In order that primeval men, or the apelike progenitors of man, should become social, they must have acquired the same instinctive feelings, which impel other animals to live in a body; and they no doubt exhibited the same general disposition. They would have felt uneasy when separated from their comrades, for whom they would have felt some degree of love; they would have warned each other of danger, and have given mutual aid in attack or defence. All this implies some degree of sympathy, fidelity, and courage. . . . Let it be borne in mind how all-important in the never-ceasing wars of savages, fidelity and courage must be. . . . Selfish and contentious people will not cohere, and without coherence nothing can be effected. A tribe rich in the above qualities would spread and be victorious over other tribes: but in the course of time it would, judging from all past history, be in its turn overcome by some other tribe still more highly endowed. Thus the social and moral qualities would tend slowly to advance and be diffused throughout the world.

  Natural selection, then, acted to select for altruism, fidelity, bonding, and those qualities that are all part and parcel of mature love. These would have been important qualities in the formation of the kinds of societies that allowed for large-scale cooperative enterprises such as agriculture, irrigation, construction projects (like grain storehouses), welfare, and courts. With the increased time it took to raise, nurture, and educate offspring, evolution had to find a way to keep the father interested in helping.

  Psychologist Martie Haselton argues that love developed as a “commitment device.” In one experiment, she asked people to think about how much they love their partner and then try to suppress thoughts of other people they find sexually attractive. She then had the same people think about how much they sexually desire that same partner and then try again to suppress thoughts about others. Thinking about someone you love was far more effective at suppressing thoughts of others than thinking about someone you lust after—even when it is the same person. Haselton argues that this is just what you’d expect from a neurochemical adaptation to create long-term commitment. Sex plays a role in strengthening commitment too. Most species of mammals engage in sex only during limited times, primarily when the female is fertile. Humans and bonobos are notable exceptions—we have sex even when it will not result in reproduction: during non-fertile times of the month, while the female is already pregnant, and after menopause. Zoologist Desmond Morris suggests that this was to give the male more reason to stick with one female. Add to that the shot of oxytocin that is released during orgasm, and you have a neurochemical recipe for men and women wanting to stay together.

  Sex is of course a powerful drive, and implicit in Haselton’s and Darwin’s arguments is that sex in effect masquerades as love—whether we see love as a cultural, psychological, spiritual, or neurochemical invention, it functioned evolutionarily as a way to ensure that the product of sexual reproduction was well cared for. At the level of society, love has become more than just looking after one’s offspring, but society itself looking after everyone’s offspring—hence schools, soccer clubs, welfare, Medicare, and courts. In other words, love for one’s partner and children evolved, culturally (and perhaps biologically), into the capacity to love life and fairness, goodness, and equality, and all the ideals we associate with society.

  Like religion and love, music itself may be similarly decomposable into the Chomskian two-part system. Many animals produce acoustic signals that sound like music to us—think birds and whales, for example. But these animal “songs” almost always lack the ability to indefinitely and infinitely recombine elements according to a hierarchy, and they lack the recursion that characterizes human musics. Interestingly, the newest research shows that while animal musics don’t have these properties so closely associated with human music, the animals can process sound in this way. In other words, nonhuman species contain the very foundational abilities that were for decades considered unique to human thought; it’s just that they haven’t (yet) learned to use them on their own. In a landmark article, Daniel Margoliash, Howard Nusbaum, and colleagues showed that European starlings can learn syntactic recursion. Earlier, Gary Rose found that white-crowned sparrows can assemble an entire song in proper sequence when exposed to only fragments of that song—suggesting that they possess an innate understanding of a syntactical rule for how sparrow songs are constructed. None of this is surprising given the dominant theme of evolution presented in this book—that its products tend to fall along a continuum.

  Animal music is purely conveyance—its purpose is to broadcast a limited number of states-of-being. Human music comprises both conveyance and rearrangement. The computational aspect enables us to plan—to contemplate how we want to use music. We use music to convey feelings and concepts we are not necessarily feeling at the present moment. We can decide to use music in order to accomplish a particular goal. I can take an element from this song and combine it with that one over there. One of my favorite songs by Rodney Crowell is a song about his favorite song. In “I Walk the Line, (Revisited),” he writes about the first time he heard the famous song by Johnny Cash:I’m back on board that ’49 Ford in 1956

  Long before the sun came up way out in the sticks

  Note Rodney’s use of internal rhymes for board and Ford (and then before in the next line), plus the alliteration of back on board, forty-nine Ford, and nineteen fifty-six. The phrases “back on board ” and “ ’49 Ford ” have the kind of percussive consonant quality that conveys an early rock ’n’ roll feel. The second line, “Long before the sun came up,” establishes the time of day, the young Crowell riding in the backseat at dawn. It also functions metaphorically, referring to this record as coming out during the dawn of rock ’n’ roll. The metaphor is even sweeter to those who know that Sun Records, through its hits with Cash, Elvis Presley, Roy Orbison, and Carl Perkins, is considered by many to be the first and most important rock ’n’ roll record label.The headlight showed a two rut roadway back up in the pines

  First time I heard Johnny Cash sing “I Walk the Line”

  In the middle of the song, Johnny himself makes a cameo, singing the famous chorus from that famous song, his booming bass voice seemingly scraping the bottom of the range of human hearing. This sort of embedding of one phrase in another is a hallmark of human language, but turns out to have correlates in other species’ vocalizations as well. Magpies and mockingbirds embed bits and pieces of other birds’ songs in their own. The point is that computational modules in the brain that give us the capacity to do this fall along a continuum across species, but reach their peak in human beings. Jazz players quote other music all the time, a trick they borrowed from the great composers, including Haydn and Mozart, who embedded pieces of other songs into their own. And to my ears, nothing sounds as sweet as Rodney’s music—and wordplay in this song, a moving and loving tribute to memory, to music, and to human creativity.

  Human music has hierarchical structure and complex syntax, and we compose within that constraint. Music, like language and religion, contains elements shared with other species and also elements unique to humans. Only humans compose a song for a particular purpose, made up of elements found in other songs. Only humans have the va
st repertoire of songs (the average American can readily identify more than one thousand different songs). Only humans have a cultural history of songs that fall within six distinct forms.

  It is important, when considering animal music, to distinguish beween musical expression and musical experience. In other words, many animals express themselves in ways that sound musical to us, but are really functioning as conveyance and messages among themselves; there is no evidence that they experience music as the aesthetic or creative artform that we do.

  It is advisable also to consider that music per se is not what evolved, but rather, music comprises components each of which has undergone a specific and probably separate evolutionary trajectory. Pitch, rhythm, and timbre are processed in separate parts of the brain. They come together later during processing, and melody, a higher-order concept, is constructed from these, influenced by changes in any one, or any combination, of the lower-level features. Music as we know it emerged in evolution after these component processes were already in place.